107
NBER WORKING PAPER SERIES BEHAVIORAL CORPORATE FINANCE: AN UPDATED SURVEY Malcolm Baker Jeffrey Wurgler Working Paper 17333 http://www.nber.org/papers/w17333 NATIONAL BUREAU OF ECONOMIC RESEARCH 1050 Massachusetts Avenue Cambridge, MA 02138 August 2011 This survey updates and extends a survey coauthored with Rick Ruback that was published in the Handbook in Corporate Finance: Empirical Corporate Finance, edited by Espen Eckbo, in 2007. We thank him for his many contributions that carried over to this version, and we thank Milt Harris for extensive and helpful comments. Baker gratefully acknowledges financial support from the Division of Research of the Harvard Business School. Malcolm Baker serve as a compensated consultant for Acadian Asset Management, a global equity manager and a director for TAL, a publicly traded lessor of containers and chassis. The views expressed herein are those of the authors and do not necessarily reflect the views of the National Bureau of Economic Research. NBER working papers are circulated for discussion and comment purposes. They have not been peer- reviewed or been subject to the review by the NBER Board of Directors that accompanies official NBER publications. © 2011 by Malcolm Baker and Jeffrey Wurgler. All rights reserved. Short sections of text, not to exceed two paragraphs, may be quoted without explicit permission provided that full credit, including © notice, is given to the source.

Behavioral Corporate Finance: An Updated Survey · Behavioral corporate finance, and behavioral finance more broadly, received a boost from the spectacular rise and fall of Internet

  • Upload
    others

  • View
    10

  • Download
    0

Embed Size (px)

Citation preview

Page 1: Behavioral Corporate Finance: An Updated Survey · Behavioral corporate finance, and behavioral finance more broadly, received a boost from the spectacular rise and fall of Internet

NBER WORKING PAPER SERIES

BEHAVIORAL CORPORATE FINANCE:AN UPDATED SURVEY

Malcolm BakerJeffrey Wurgler

Working Paper 17333http://www.nber.org/papers/w17333

NATIONAL BUREAU OF ECONOMIC RESEARCH1050 Massachusetts Avenue

Cambridge, MA 02138August 2011

This survey updates and extends a survey coauthored with Rick Ruback that was published in the Handbookin Corporate Finance: Empirical Corporate Finance, edited by Espen Eckbo, in 2007. We thank himfor his many contributions that carried over to this version, and we thank Milt Harris for extensiveand helpful comments. Baker gratefully acknowledges financial support from the Division of Researchof the Harvard Business School. Malcolm Baker serve as a compensated consultant for Acadian AssetManagement, a global equity manager and a director for TAL, a publicly traded lessor of containersand chassis. The views expressed herein are those of the authors and do not necessarily reflect theviews of the National Bureau of Economic Research.

NBER working papers are circulated for discussion and comment purposes. They have not been peer-reviewed or been subject to the review by the NBER Board of Directors that accompanies officialNBER publications.

© 2011 by Malcolm Baker and Jeffrey Wurgler. All rights reserved. Short sections of text, not to exceedtwo paragraphs, may be quoted without explicit permission provided that full credit, including © notice,is given to the source.

Page 2: Behavioral Corporate Finance: An Updated Survey · Behavioral corporate finance, and behavioral finance more broadly, received a boost from the spectacular rise and fall of Internet

Behavioral Corporate Finance: An Updated SurveyMalcolm Baker and Jeffrey WurglerNBER Working Paper No. 17333August 2011JEL No. G3,G30,G31,G32,G34,G35

ABSTRACT

We survey the theory and evidence of behavioral corporate finance, which generally takes one of twoapproaches. The market timing and catering approach views managerial financing and investmentdecisions as rational managerial responses to securities mispricing. The managerial biases approachstudies the direct effects of managers’ biases and nonstandard preferences on their decisions. We reviewrelevant psychology, economic theory and predictions, empirical challenges, empirical evidence, newdirections such as behavioral signaling, and open questions.

Malcolm BakerBaker Library 261Harvard Business SchoolSoldiers FieldBoston, MA 02163and [email protected]

Jeffrey WurglerStern School of Business, Suite 9-190New York University44 West 4th StreetNew York, NY 10012and [email protected]

Page 3: Behavioral Corporate Finance: An Updated Survey · Behavioral corporate finance, and behavioral finance more broadly, received a boost from the spectacular rise and fall of Internet

Table of Contents 

1.   Introduction ................................................................................................................................... 1 

2.   Market timing and catering ...................................................................................................... 5 2.1.   Background on investor behavior and market inefficiency ................................................. 5 2.1.1.  Limited arbitrage ............................................................................................................................................ 6 2.1.2.  Categorization and investor sentiment ................................................................................................. 8 2.1.3.  Prospect theory, reference points, loss aversion, and anchoring ............................................... 9 2.1.4.  Smart managers ............................................................................................................................................ 11 

2.2.  Theoretical framework: Rational managers in irrational markets ................................. 12 2.3.  Empirical challenges ......................................................................................................................... 18 2.4.  Investment policy ............................................................................................................................... 22 2.4.1.   Real investment ............................................................................................................................................. 22 2.4.2.   Mergers and acquisitions .......................................................................................................................... 25 2.4.3.   Diversification and focus ........................................................................................................................... 27 

2.5.   Financial policy .................................................................................................................................. 28 2.5.1.   Equity issues ................................................................................................................................................... 28 2.5.2.   Repurchases .................................................................................................................................................... 33 2.5.3.   Debt issues ....................................................................................................................................................... 34 2.5.4.   Cross‐border issues ..................................................................................................................................... 37 2.5.5.   Financial intermediation ........................................................................................................................... 37 2.5.6.   Capital structure ........................................................................................................................................... 41 

2.6.   Other corporate decisions ............................................................................................................. 43 2.6.1.   Dividends ......................................................................................................................................................... 43 2.6.2.   Earnings management ................................................................................................................................ 45 2.6.3.   Firm names...................................................................................................................................................... 47 2.6.4.  Nominal share prices .................................................................................................................................. 48 2.6.5.   Executive compensation ............................................................................................................................ 49 

3.   Managerial Biases ...................................................................................................................... 50 3.1.  Background on managerial behavior ......................................................................................... 50 3.1.1.   Limited governance ..................................................................................................................................... 50 3.1.2.  Bounded rationality ..................................................................................................................................... 51 3.1.3.  Optimism, overconfidence and hubris ................................................................................................. 52 3.1.4.  More on reference dependence .............................................................................................................. 53 

3.2.   Theoretical framework ................................................................................................................... 54 3.3.   Empirical challenges ........................................................................................................................ 57 3.4.   Investment policy .............................................................................................................................. 58 3.4.1.   Real investment ............................................................................................................................................. 58 3.4.2.   Mergers and acquisitions .......................................................................................................................... 62 

3.5.   Financial policy .................................................................................................................................. 65 3.5.1.  Equity issues ................................................................................................................................................... 65 3.5.2.  IPO prices ......................................................................................................................................................... 66 

Page 4: Behavioral Corporate Finance: An Updated Survey · Behavioral corporate finance, and behavioral finance more broadly, received a boost from the spectacular rise and fall of Internet

3.5.3.   Raising debt .................................................................................................................................................... 67 3.5.4.   Capital structure ........................................................................................................................................... 68 3.5.5.   Contracting and executive compensation .......................................................................................... 69 

4.   Behavioral Signaling ................................................................................................................. 71 4.1.  Theoretical Framework ................................................................................................................... 72 4.2.   Applications ........................................................................................................................................ 77 4.2.1  Dividends ......................................................................................................................................................... 77 4.2.2.   Other applications ........................................................................................................................................ 78 

5.   Some open questions ................................................................................................................ 80  

Page 5: Behavioral Corporate Finance: An Updated Survey · Behavioral corporate finance, and behavioral finance more broadly, received a boost from the spectacular rise and fall of Internet

  1 

1.   Introduction 

Corporate finance aims to explain the financial contracts and the real investment behavior 

that emerge from the interaction of managers and investors. A complete explanation of financing 

and investment patterns therefore requires a correct understanding of the beliefs and preferences 

of these two sets of agents. The majority of research in corporate finance makes broad assumptions 

that these beliefs and preferences are fully rational. Agents are supposed to develop unbiased 

forecasts about future events and use these to make decisions that best serve their own interests. 

As a practical matter, this means that managers can take for granted that capital markets are 

efficient, with prices rationally reflecting public information about fundamental values. Likewise, 

investors can take for granted that managers will act in their self‐interest, rationally responding to 

incentives shaped by compensation contracts, the market for corporate control, and other 

governance mechanisms. 

Research in behavioral corporate finance replaces the traditional rationality assumptions 

with behavioral foundations that are more evidence‐driven. The field is no longer a purely 

academic pursuit, as behavioral corporate finance is increasingly the basis of discussions in 

mainstream textbooks.1  We divide the literature into two broad groups and organize the survey 

accordingly. Roughly speaking, the first approach emphasizes the effect of investor behavior that is 

less than fully rational. The second considers managerial behavior that is less than fully rational. 

For each line of research, we review the basic theoretical frameworks, the main empirical 

challenges, and the evidence. Of course, in practice, multiple channels of irrationality may operate 

at the same time; our taxonomy is meant to fit the bulk of the existing literature.  

                                                             

1 For example see Damodaran (2011), Shefrin (2006), Shefrin (2008), and Welch (2009). 

Page 6: Behavioral Corporate Finance: An Updated Survey · Behavioral corporate finance, and behavioral finance more broadly, received a boost from the spectacular rise and fall of Internet

  2 

The “market timing and catering approach” assumes that arbitrage in securities markets is 

imperfect, and as a result prices can be too high or too low. We review the market inefficiency 

literature insofar as it is relevant. Rational managers are assumed to perceive these mispricings, 

and to make decisions that exploit or further encourage mispricing. While their decisions may 

maximize the short‐run value of the firm, they may also result in lower long‐run values as prices 

correct to fundamentals. In the simple theoretical framework we outline, managers balance three 

objectives: fundamental value, catering, and market timing. Maximizing fundamental value has the 

usual ingredients. Catering refers to any actions intended to boost share prices above fundamental 

value. Market timing refers to financing decisions intended to capitalize on temporary mispricings, 

generally by issuing overvalued securities and repurchasing undervalued ones.  

Empirical tests of the irrational investors model face the challenge of measuring mispricing. 

We discuss how this issue has been tackled. A few papers use clever approaches that can identify 

mispricing fairly convincingly, but in many cases ambiguities remain. Overall, despite some 

unresolved questions, the evidence suggests that the irrational investors approach has a 

considerable degree of descriptive power. We review studies on investment behavior, merger 

activity, the clustering and timing of corporate security offerings, capital structure, corporate name 

changes, nominal share prices, dividend policy, earnings management, and other managerial 

decisions. We also point out gaps that remain between the theory and the evidence.  

The second approach that we discuss is the “managerial biases” approach. It assumes that 

managers have behavioral biases, but retains the rationality of investors, albeit limiting the 

governance mechanisms they can employ to constrain managers. Following the emphases of the 

current literature, our discussion centers on the biases of optimism and overconfidence. A simple 

model shows how these biases, in leading managers to believe their firms are undervalued, 

encourage overinvestment from internal resources, and a preference for internal to external 

Page 7: Behavioral Corporate Finance: An Updated Survey · Behavioral corporate finance, and behavioral finance more broadly, received a boost from the spectacular rise and fall of Internet

  3 

finance, especially internal equity. We note that the predictions of the optimism and overconfidence 

models typically look very much like those of agency and asymmetric information models.  

In this approach, the main obstacles for empirical tests include distinguishing predictions 

from standard, non‐behavioral models, as well as empirically measuring managerial biases. Again, 

however, creative solutions have been proposed. The effects of optimism and overconfidence have 

been empirically studied in the context of corporate and entrepreneurial financing and investment 

decisions, merger activity, and the structure of financial contracts.  

We also cover a newer approach that we call “behavioral signaling.” This is a response to the 

many sophisticated signaling models in corporate finance theory that make two questionable 

assumptions. They assume full rationality and standard preferences; and, they use the destruction 

of firm value as the credible signaling mechanism—the better firm is the one that destroys more 

value, a notion rejected by managers in surveys. Behavioral signaling models instead base the 

signaling mechanism on some distortion in beliefs or preferences. We describe a model of 

dividends where investors are loss‐averse over the level of dividends, so that a manager that 

ratchets up dividends today can signal that he can likely meet or exceed that level tomorrow. 

Following this, we speculate about other topics that might be addressed when asymmetric 

information is combined with nonstandard preferences or biased expectations.  

Sprinkled throughout the survey are discussions of research that is hard to categorize into 

just one paradigm. For example, mergers are arranged by bankers and two sets of managers and 

approved by shareholders; behavioral biases that affect the outcome are difficult to attribute to one 

party. They may well be shared across parties. Complications like these suggest why the real 

economic losses associated with behavioral phenomena in corporate finance are hard to quantify, 

although some evidence suggests that they are considerable. 

Page 8: Behavioral Corporate Finance: An Updated Survey · Behavioral corporate finance, and behavioral finance more broadly, received a boost from the spectacular rise and fall of Internet

  4 

Behavioral corporate finance, and behavioral finance more broadly, received a boost from 

the spectacular rise and fall of Internet stocks between the mid‐1990s and 2000. It is hard to 

explain this period, both at the level of market aggregates and individual stocks and other 

securities, without appealing to some degree of investor and managerial irrationality.  

The more recent financial crisis is more complex, as we discuss. The mispricing did not 

involve a new technology, but rather more mundane mortgage finance made opaque through 

financial innovation and the creation of seemingly low‐risk derivatives. The buyers were not retail 

investors, but banks and money market mutual funds. Most importantly, the systemically important 

banks that created these securities had some of the largest exposures. It was as if Bank of America 

had held on to a large fraction of the Internet stocks that were underwritten in the late 1990s. 

There were equal parts traditional corporate finance frictions, like agency problems, signaling, and 

debt overhang, and behavioral distortions that led to both the credit bubble and the challenges of 

resetting bank balance sheets. The economic damage was further multiplied because banks 

themselves shouldered the losses.  

Taking a step back, it is important to note that the approaches take very different views 

about the role and quality of managers, and have very different normative implications as a result. 

For example, when the primary source of irrationality is on the investor side, as in the market 

timing and catering approach and in our implementation of behavioral signaling, long‐term value 

maximization and economic efficiency requires insulating managers from short‐term share price 

pressures. Managers need the flexibility necessary to make decisions that may be unpopular in the 

marketplace. This may imply benefits from internal capital markets, barriers to takeovers, and so 

forth—many of the institutions that are disdained by an agency perspective. On the other hand, if 

the main source of irrationality is manifested through managerial biases, efficiency requires 

Page 9: Behavioral Corporate Finance: An Updated Survey · Behavioral corporate finance, and behavioral finance more broadly, received a boost from the spectacular rise and fall of Internet

  5 

reducing discretion and obligating managers to respond to market price signals—as standard 

agency theory and asymmetric information models would have it.  

The stark contrast between the normative implications of different approaches to 

behavioral corporate finance is one reason why the area is fascinating, and why more work in the 

area may lead to important insights. Our ever‐improving understanding of the economic 

implications of social psychology and the ever‐increasing availability of micro data will continue to 

present new research opportunities. In that vein, we close the survey with some open questions.  

And at this point we would also like to point the reader to excellent recent surveys of 

individual topics in behavioral corporate finance: Ben‐David (2010) on dividend policy, Derrien 

(2010) on IPOs, Dong (2010) on mergers and acquisitions, Gider and Hackbarth (2010) on 

financing decisions, Gervais (2010) on investment decisions, and Morck (2010) on governance.  

2.   Market timing and catering 

The most developed framework in behavioral corporate finance and longest section in this 

survey involves rational managers interacting with irrational investors.  

2.1.   Background on investor behavior and market inefficiency 

There are two key building blocks in the market timing and catering framework. The first is 

that irrational investors must influence securities prices. In other words, that securities markets are 

not entirely informationally efficient. Otherwise, it is not obvious that managers would take much 

care to please such investors. For irrational investors to affect prices, rational investors must be 

limited in their ability to compete and arbitrage away mispricings. We discuss the limited arbitrage 

literature below since this is such a critical assumption.  

Page 10: Behavioral Corporate Finance: An Updated Survey · Behavioral corporate finance, and behavioral finance more broadly, received a boost from the spectacular rise and fall of Internet

  6 

Irrational traders’ biases must be systematic, as well, or else their own trading might simply 

cancel out, leaving arbitrageurs with little to do anyway. We discuss a few well‐documented and 

robust deviations from standard utility and Bayesian beliefs from the psychology, economics, and 

finance literatures. The particular deviations that are most immediately applicable to corporate 

finance involve categorization and reference‐dependent behavior. Combined with limited arbitrage, 

these biases lead to market inefficiencies.2  

The second key building block of the market timing and catering view is that managers 

must be “smart” in the sense of being able to distinguish market prices and fundamental value—to 

recognize the mispricings that irrational investors have created, especially in extreme 

circumstances. We review several reasons why this assumption is plausible.   

2.1.1.  Limited arbitrage 

  Securities prices reflect fundamental values when informed investors compete aggressively 

to eliminate mispricings. Classical finance theory, including the Modigliani‐Miller theorem, holds 

that they will do so because mispricings between two companies with the same operating cash 

flows but different capital structures, in a setting of complete and frictionless securities markets, 

present arbitrage opportunities. The assumption of market efficiency has for decades permitted 

corporate finance theory to develop independently of asset pricing theory.  

                                                             

2 The literature on market inefficiency is vast. It includes fairly convincing evidence of inefficiencies including the January effect; the effect of trading hours on price volatility; post‐earnings‐announcement drift, positive autocorrelation in quarterly earnings announcement effects, and more generally delayed reaction to news; momentum; Siamese twin securities that have identical cash flows but trade at different prices; negative “stub” values; closed‐end fund pricing patterns; bubbles and crashes in growth stocks; related evidence of mispricing in options, bond, and foreign exchange markets; and interesting new patterns every year. This list excludes anomalies related to securities issuance that we discuss later. See Barberis and Thaler (2003) and Shleifer (2000) for classic surveys of the behavioral finance and asset pricing literature more broadly. 

Page 11: Behavioral Corporate Finance: An Updated Survey · Behavioral corporate finance, and behavioral finance more broadly, received a boost from the spectacular rise and fall of Internet

  7 

The literature on limited arbitrage, however, concludes that securities market mispricings 

often do not present opportunities for true arbitrage. As a result, mispricings can exist and persist. 

As just one example, the fact that stocks added to market indexes see their prices jump has been 

viewed as prima facie proof of limits to arbitrage in the stock market (Shleifer (1986) and Harris 

and Gurel (1986)). A deeper study of specific arbitrage costs and risks is useful, however, because 

when these costs are measurable, they may lead to empirical strategies for measuring mispricing, 

as we discuss later.   

Early contributions to the literature include Miller (1977), who points out that short‐sale 

constraints can lead to securities being overpriced. DeLong, Shleifer, Summers, and Waldmann 

(1990) highlight the risk that irrational traders push prices further away from fundamentals after a 

would‐be arbitrageur takes a position. Shleifer and Vishny (1997) point out that professional 

investment managers, the enforcers of market efficiency in classical theory, in fact have a special 

incentive to avoid this noise trader risk: in the realistic case where investors cannot distinguish 

between returns earned by luck and skill, they may assume the worst and withdraw funds when 

faced with losses. 

  There are a number of additional costs and risks of arbitrage. An important one is 

fundamental risk, which makes relative‐value arbitrage risky because a mispriced security’s cash 

flows are not spanned by those of other assets (Pontiff (1996) and Wurgler and Zhuravskaya 

(2002)). Liquidity risk arises when everyone wants to sell at the same time (Acharya and Pedersen 

(2004)). Finally, real‐world investors must bear simple transaction costs, search costs, and 

information‐gathering costs to exploit mispricings.  

  The idea that securities prices are affected by more than just fundamentals has been 

examined in markets from penny stocks to government bonds. Krishnamurthy (2002) finds that on‐

the‐run Treasury issues trade at a premium to other bonds, while Duffee (1996) connects the 

Page 12: Behavioral Corporate Finance: An Updated Survey · Behavioral corporate finance, and behavioral finance more broadly, received a boost from the spectacular rise and fall of Internet

  8 

supply of individual bills to non‐fundamental variation in the Treasury yield curve. At a higher level 

of aggregation, Hu, Pan, and Wang (2010) use anomalous patterns in the shape of the yield curve to 

quantify how well capitalized or effective is bond market arbitrage. At the broadest level, 

Greenwood and Vayanos (2010) argue that the overall shape of the yield curve is causally affected 

by the maturity structure of government debt issues. This assertion implies mispricings of far 

greater size than those evidenced by relative‐value distortions within the yield curve—large 

enough, perhaps, to catch the attention of managers, or their investment bankers, and affect 

corporate maturity structure choices.  

  In summary, a body of theory and evidence indicates that capital markets have a limited 

capacity to absorb demand shocks that are independent of fundamental news. The next task is to 

understand the investor psychology that is behind some of these demand shocks.  

2.1.2.  Categorization and investor sentiment 

  A basic feature of human cognition is simplification through categories. For example, the 

label “Behavioral Corporate Finance” defines a set of papers with similar methodological themes 

and frees us from having to enumerate the individual members of the set (except in the case of a 

survey article, of course). The classic treatment is Rosch (1973), but the principle is obvious and 

needs no theoretical preamble.  

Investors and analysts simplify the investment universe through categories (Barberis and 

Shleifer (2003)). Some categories, such as small‐caps, value stocks, high‐yield stocks, and junk 

bonds, are fairly timeless. Others are ephemeral. The “Nifty Fifty” is a forgotten moniker from the 

early 1970s for a set of large‐capitalization firms with solid earnings growth. These days, “Internet 

firms” is becoming a less useful label. It once denoted firms with the essential feature that their 

success depended on the adoption of a new technology; that technology is now established, so the 

determinants of these firms’ prospects have become more individualized.  

Page 13: Behavioral Corporate Finance: An Updated Survey · Behavioral corporate finance, and behavioral finance more broadly, received a boost from the spectacular rise and fall of Internet

  9 

  Investment categories become interesting for us when investors trade at the category level. 

Index funds provide an example of category‐level investing and its consequences: When a stock is 

added to the S&P 500 Index, its returns become more correlated with existing Index members 

(Barberis, Shleifer, and Wurgler (2005)). It is now traded in sync with them, and—arbitrage being 

limited—it acquires a common factor in returns. Over time, this can lead to a detachment of 

category members from the rest of the market (Morck and Yang (2001), Wurgler (2011)). The most 

dramatic cases are from bubbles and crashes. In the Internet bubble, some investors didn’t have the 

time or expertise to investigate individual tech stocks and apparently just threw money at anything 

Internet‐related. The crash involved equally indiscriminate selling. A qualitative review of stock 

market history suggests that investor sentiment often concentrates at the level of categories.  

For our purpose, categorization will be particularly relevant to the discussion of catering 

behavior, in which managers take actions to move their firm into the in‐vogue category and boost 

its valuation. This boost may, in turn, facilitate opportunistic securities issuance.  

2.1.3.  Prospect theory, reference points, loss aversion, and anchoring  

  In the prospect theory preferences of Kahneman and Tversky (1979), utility is defined not 

as a smoothly increasing function of the level of consumption or wealth but in terms of changes 

relative to a reference level. Via a kink at the origin, the value function also embodies loss 

aversion—the empirical phenomenon that losses, even small ones, are particularly painful. See 

Tversky and Kahneman (1991) for a survey of loss aversion research. 

  The disposition effect of Shefrin and Statman (1985) refers to the pattern that investors are 

more likely to realize gains than losses. A typical explanation invokes elements of prospect theory: 

the reference point is the purchase price, and the investor strains to avoid selling at a loss despite 

Page 14: Behavioral Corporate Finance: An Updated Survey · Behavioral corporate finance, and behavioral finance more broadly, received a boost from the spectacular rise and fall of Internet

  10 

the tax advantage to doing so.3 Other salient reference prices, and, importantly, ones that are 

common across investors, are recent high prices, such as a stock’s all‐time or 52‐week high, and 

recent low prices. Huddart, Lang, and Yetman (2009) find that trading volume and return patterns 

change as recent highs are approached for seasoned issues, and Kaustia (2004) finds that trading 

volume behavior changes as IPOs reach new maxima and minima.  

  Tversky and Kahneman (1974) also review the concept of anchoring. Anchoring refers to a 

deviation from Bayesian beliefs, not a departure from standard preferences. In anchoring, the 

subject forms beliefs by adjusting from a potentially arbitrary starting point, and the bias is that the 

final belief is biased toward this anchor; adjustment away from it is insufficient. For example, 

Tversky and Kahneman asked subjects to guess what fraction of African countries were members of 

the United Nations. Those who were first asked “is it more or less than 10%?” guessed a median of 

25%, while those who had been asked ”is it more or less than 65%” guessed a median of 45%.  

Offering payoffs for accuracy did not reduce these effects. Another example comes from Strack and 

Mussweiler (1997), who asked subjects to estimate when Einstein first visited the United States. 

Implausible anchors like 1215 and 1992 produced effects as large as anchors of 1909 and 1939. 

Studies involving reference point thinking, loss aversion, and anchoring are featured at 

several points in this survey. These phenomena have been used to shed light on dividends, earnings 

management, merger offer prices, equity issuance timing, hurdle rates, the cost of debt, and other 

patterns.  

                                                             

3 Barberis and Xiong (2009) and Kaustia (2010b) show that empirical features of the disposition effect make it hard to connect to prospect theory per se, which also specifies curvature in the value function. See Kaustia (2010a) for a thorough survey of the disposition effect literature. 

Page 15: Behavioral Corporate Finance: An Updated Survey · Behavioral corporate finance, and behavioral finance more broadly, received a boost from the spectacular rise and fall of Internet

  11 

2.1.4.  Smart managers 

But even if limited arbitrage and systematic investor biases add up to inefficient markets, 

why is it reasonable to assume that corporate managers are “smart” in the sense of being able to 

identify mispricing? One can offer several justifications. First, corporate managers have superior 

information about their own firm. This is evidenced by the abnormally high returns on illegal 

insider trading in Muelbroek (1992) and even legal insider trading in Seyhun (1992).  

Second, managers can manufacture their own information advantage by managing earnings 

or with the help of conflicted analysts, as in Bradshaw, Richardson, and Sloan (2006). They may 

also be able to shape investor demand through investor relations, by marketing their shares in Gao 

and Ritter (2010), or allocating IPO shares in Zhang (2004). 

Third, corporate managers have fewer constraints than equally “smart” money managers. 

Consider two classic models of limited arbitrage introduced above: DeLong et al. (1990) is built on 

short horizons and Miller (1977) on short‐sales constraints. CFOs tend to be judged on longer 

horizon results than are money managers, allowing them to take a view on market valuations in a 

way that most money managers cannot.4 Short‐sales constraints also prevent money managers 

from mimicking CFOs. When a firm or a sector becomes overvalued, corporations are the natural 

candidates to expand the supply of shares.5 Money managers are not. 

In addition, managers might just follow intuitive rules of thumb that allow them to identify 

mispricing even without any real information advantage. In Baker and Stein (2004), one such 

                                                             

4 For example, suppose the manager issues equity at $50 per share. Should those shares subsequently double, the manager might regret not delaying the issue, but he will surely not be fired, having presided over a rise in the stock price. In contrast, imagine a money manager sells (short) the same stock at $50. This might lead to considerable losses for the firm and the executive, an outflow of funds, and, if the bet is large enough, perhaps the end of a career. 

5 Conversely, when the shares crash, firms serve as buyers of last resort (Hong, Wang, and Yu (2008)).  

Page 16: Behavioral Corporate Finance: An Updated Survey · Behavioral corporate finance, and behavioral finance more broadly, received a boost from the spectacular rise and fall of Internet

  12 

successful rule of thumb is to issue equity when the market is particularly liquid, in the sense of a 

small price impact upon the issue announcement. In the presence of short‐sales constraints—more 

on this below—unusually high liquidity is symptomatic of an overvalued market dominated by 

irrationally optimistic investors. 

Finally, in the case of debt maturity, firms may have a comparative advantage in exploiting 

distortions in the yield curve. Greenwood, Hanson, and Stein (2010) develop this logic. In a 

Modigliani‐Miller world, firms are indifferent to their debt maturity, freeing them to fill in the gap in 

supply at various maturities created by restructuring of the Treasury debt maturity structure or 

other non‐fundamental supply and demand effects on the yield curve. By contrast, mutual fund and 

institutional investment managers often have less flexibility, by mandate and other limits of 

arbitrage, to be opportunistic in their maturity choice.  

2.2.  Theoretical framework: Rational managers in irrational markets 

We use the assumptions of inefficient markets and smart managers to develop a simple 

theoretical framework for the market timing and catering approach. The framework has roots in 

Fischer and Merton (1984), De Long, Shleifer, Summers, and Waldmann (1989), Morck, Shleifer, 

and Vishny (1990b), and Blanchard, Rhee, and Summers (1993), but our particular derivation 

borrows most from Stein (1996). Newer models, such as Bolton, Chen, and Wang (2011), add 

dynamic considerations to this static framework.  

In the market timing and catering approach, the manager balances three conflicting goals. 

The first is to maximize fundamental value. This means selecting and financing investment projects 

to increase the rationally risk‐adjusted present value of future cash flows. To simplify the analysis, 

we do not explicitly model taxes, costs of financial distress, agency problems or asymmetric 

information. Instead, we specify fundamental value as  

Page 17: Behavioral Corporate Finance: An Updated Survey · Behavioral corporate finance, and behavioral finance more broadly, received a boost from the spectacular rise and fall of Internet

  13 

KKf , , 

where f is increasing and concave in new investment K. To the extent that any of the usual market 

imperfections leads the Modigliani‐Miller (1958) theorem to fail, financing may enter f alongside 

investment. 

The second goal is to maximize the current share price of the firm’s securities. In perfect 

capital markets, the first two objectives are the same, since the definition of market efficiency is 

that price equals fundamental value. But once one relaxes the assumption of investor rationality, 

this need not be true, and the second objective is distinct. In particular, the second goal is to “cater” 

to short‐term investor demands via particular investment projects or otherwise packaging the firm 

and its securities in a way that maximizes appeal to investors. Through such catering activities, 

managers influence the temporary mispricing, which we represent by the function 

where the arguments of  depend on the nature of prevailing investor sentiment. The arguments 

might include investing in a particular technology, assuming a conglomerate or single‐segment 

structure, changing the corporate name, managing earnings, initiating a dividend, splitting shares, 

and so on. In practice, the determinants of mispricing may well vary over time.  

The third goal is to exploit the current mispricing for the benefit of existing, long‐run 

investors. Managers achieve this by a “market timing” financing policy which supplies securities 

that are temporarily overvalued and repurchases those that are undervalued, or at least less 

overvalued. This policy transfers value from the new or the outgoing investors to the ongoing, long‐

Page 18: Behavioral Corporate Finance: An Updated Survey · Behavioral corporate finance, and behavioral finance more broadly, received a boost from the spectacular rise and fall of Internet

  14 

run investors; the transfer is realized as prices correct in the long run.6 For simplicity, we focus 

here on temporary mispricing in the equity markets, and so  refers to the difference between the 

current price and the fundamental value of equity. More generally, each of the firm’s securities may 

be mispriced to some degree. By selling a fraction of the firm e, long run shareholders gain 

e .7 

We leave out the budget constraint and lump together the sale of new and existing shares. Instead 

of explicitly modeling the flow of funds and any potential financial constraints, we will consider the 

reduced form impact of e on fundamental value.  

It is worth noting that other capital market imperfections can lead to a sort of catering 

behavior. For example, reputation models in the spirit of Holmstrom (1982) can lead to earnings 

management, inefficient investment, and excessive swings in corporate strategy even when the 

capital markets are not fooled in equilibrium.8 Viewed in this light, the framework here is relaxing 

the assumptions of rational expectations in Holmstrom, in the case of catering, and Myers and 

Majluf (1984), in the case of market timing.  

                                                             

6 Of course, we are also using the market inefficiency assumption here in assuming that managerial efforts to capture a mispricing do not fully and instantly destroy it in the process, as they do in the rational expectations world of Myers and Majluf (1984). In other words, investors underreact to corporate decisions designed to exploit mispricing because of limited arbitrage, attention, etc. 

7 For long run shareholders to benefit, we are implicitly thinking of something like three‐period model. In the first period, investment and financing decisions are made, and prices are above fundamental value by an amount . There is an intermediate period where prices do not change, but short‐run investors sell their shares, and a final period where fundamental value is realized. Issuing equity will the effect of reducing prices in the first and second periods if e < 0, while increasing the value per share in the third period from where it would otherwise be. 

8 For examples, see Stein (1989) and Scharfstein and Stein (1990). For a comparison of rational expectations and inefficient markets in this framework, see Aghion and Stein (2008). 

Page 19: Behavioral Corporate Finance: An Updated Survey · Behavioral corporate finance, and behavioral finance more broadly, received a boost from the spectacular rise and fall of Internet

  15 

Putting the goals of fundamental value, catering, and market timing into one objective 

function, we have the manager choosing investment and financing to 

1,max,

eKKfeK

where  is greater than zero and less than or equal to one and specifies the manager’s horizon. 

When  equals one, the manager cares only about creating value for existing, long‐run 

shareholders, the last term drops out, and there is no distinct impact of catering. However, and 

interestingly, even an extremely long‐horizon manager cares about short‐term mispricing for the 

purposes of market timing, and thus may cater to short‐term mispricing to further this objective. 

With a shorter horizon, maximizing the stock price becomes an objective in its own right, even 

without any concomitant equity issues.  

We take the managerial horizon as exogenously set by personal characteristics, career 

concerns, and the compensation contract. If the manager plans to sell equity or exercise options in 

the near term, his portfolio considerations may lower . Career concerns and the market for 

corporate control can also combine to shorten horizons: if the manager does not maximize short‐

run prices, the firm may be acquired and the manager fired. 

Differentiating with respect to K and e gives the optimal investment and financial policy of a 

rational manager operating in inefficient capital markets: 

KK eKf

11, , and 

ee eKf

1, . 

The first condition is about investment policy. The marginal value created from investment 

is weighed against the standard cost of capital, normalized to be one here, net of the impact that 

Page 20: Behavioral Corporate Finance: An Updated Survey · Behavioral corporate finance, and behavioral finance more broadly, received a boost from the spectacular rise and fall of Internet

  16 

this incremental investment has on mispricing, and hence its effect through mispricing on catering 

and market timing gains. The second condition is about financing. The marginal value lost from 

shifting the firm’s current capital structure toward equity is weighed against the direct market 

timing gains and the impact that this incremental equity issuance has on mispricing, and hence its 

effect on catering and market timing gains. This is a lot to swallow at once, so we consider some 

special cases. 

Investment policy. Investment and financing are separable if both K and feK are equal to 

zero. Then the investment decision reduces to the familiar perfect markets condition of fK equal to 

unity. Note that, if fe is equal to zero, there is no optimal capital structure. Real consequences of 

mispricing for investment arise in two ways. Either capital structure has a real effect on value, when 

ef  and  eKf  are not equal to zero, or investment has a direct effect on mispricing, when  K  is not 

equal to zero. The simplest situation to evaluate in the first case has  K  and  e  equal to zero. The 

simplest situation to evaluate in the second case is when  ef  is equal to zero. Both channels are 

likely present, but analyzing the two at the same time reduces transparency. 

In Stein (1996) and Baker, Stein, and Wurgler (2003), fe and feK are not equal to zero. There 

is an optimal capital structure, or at least an upper bound on debt capacity. The benefits of issuing 

or repurchasing equity in response to mispricing are balanced against the reduction in fundamental 

value that arises from too much (or possibly too little) leverage and the indirect effect on firm value 

through investment, when feK is greater than zero. Somewhat more formally, equity issues e are 

increasing in an exogenous level of mispricing . (This also requires the assumption that fee is less 

than zero, which is necessary for an interior solution for optimal capital structure.) To match Baker, 

Stein and Wurgler (2003), consider the case of an undervalued firm. The more undervalued the 

firm, the less equity the manager sells. This constrains investment when feK is greater than zero, i.e. 

K is increasing in e. (Constraints of this type also require the assumption that fKK is less than zero, 

Page 21: Behavioral Corporate Finance: An Updated Survey · Behavioral corporate finance, and behavioral finance more broadly, received a boost from the spectacular rise and fall of Internet

  17 

which is necessary for an interior solution for investment.) In sum, because of undervaluation and 

financial constraints, the manager chooses a combination of lower equity issues e and lower 

investment K than he would in the situation of no mispricing. 

In Polk and Sapienza (2009) and Gilchrist, Himmelberg, and Huberman (2005), there is no 

optimal capital structure, but K is not equal to zero: mispricing is itself a function of investment. 

The potential to create mispricing distorts investment in a simple, direct way. Polk and Sapienza 

focus on catering effects and do not consider financing (e equal to zero in this setup), while Gilchrist 

et al. model the market timing decisions of managers with long horizons (equal to one). 

Financial policy. The demand curve for a firm’s equity slopes down under the natural 

assumption that e is negative, e.g., issuing shares partly corrects mispricing.9 When investment and 

financing are separable, managers act like monopolists. This is easiest to see when managers have 

long horizons, and they sell down the demand curve until marginal revenue  is equal to marginal 

cost –ee. Note that price remains above fundamental value even after the issue: “corporate 

arbitrage” moves the market toward, but not all the way to, market efficiency.10 Managers sell less 

equity when they care about short‐run stock price ( less than one, here). For example, in 

Ljungqvist, Nanda, and Singh (2005), managers expect to sell their own shares soon after the IPO 

and so issue less as a result. Managers also sell less equity when there are costs of suboptimal 

leverage. To some extent, the shape of the demand curve may be endogenous. Gao and Ritter 

(2010) argue that firms actively market their shares in anticipation of an equity offering with this in 

mind. 

                                                             

9 Gilchrist et al. (2005) model this explicitly with heterogeneous investor beliefs and short‐sales constraints. See also Hong, Wang, and Yu (2008)). 

10 Total market timing gains may be even higher in a dynamic model where managers can sell in small increments down the demand curve. 

Page 22: Behavioral Corporate Finance: An Updated Survey · Behavioral corporate finance, and behavioral finance more broadly, received a boost from the spectacular rise and fall of Internet

  18 

Other corporate decisions. This framework can be expanded to accommodate decisions 

beyond investment and issuance. Consider dividend policy. Increasing or initiating a dividend may 

simultaneously affect both fundamental value, through taxes, and the degree of mispricing, if 

investors categorize stocks according to payout policy as they do in Baker and Wurgler (2004a). 

The tradeoff is  

dd eKf

1, , 

where the left‐hand side is the tax cost of dividends, for example, and the right‐hand side is the 

market timing gain, if the firm is simultaneously issuing equity, plus the catering gain, if the 

manager has short horizons. In principle, a similar tradeoff governs the earnings management 

decision or corporate name changes; however, particularly in the latter case, the fundamental costs 

of catering would presumably be small. 

2.3.  Empirical challenges 

The market timing and catering framework features the role of securities mispricing in 

investment, financing, and other corporate decisions. The main challenge for empirical tests in this 

area is measuring mispricing, which by its nature is hard to pin down. Researchers have 

operationalized empirical tests in a few different ways. 

Ex ante misvaluation. One option is to take an ex ante measure of mispricing, for instance a 

scaled‐price ratio in which a market value in the numerator is related to some measure of 

fundamental value in the denominator. Perhaps the most common choice is the market‐to‐book 

ratio: A high market‐to‐book suggests that the firm may be overvalued. Consistent with this idea, 

and the presumption that mispricing corrects in the long run, market‐to‐book is found to be 

inversely related to future stock returns in the cross‐section by Fama and French (1992) and in the 

time‐series by Kothari and Shanken (1997) and Pontiff and Schall (1998). Also, extreme values of 

Page 23: Behavioral Corporate Finance: An Updated Survey · Behavioral corporate finance, and behavioral finance more broadly, received a boost from the spectacular rise and fall of Internet

  19 

market‐to‐book are connected to extreme investor expectations by Lakonishok, Shleifer and Vishny 

(1994), La Porta (1996), and La Porta, Lakonishok, Shleifer, and Vishny (1997).  

One difficulty that arises with this approach is that the market‐to‐book ratio or another ex 

ante measure of mispricing may be correlated with an array of firm characteristics. Book value is 

not a precise estimate of fundamental value, but rather a summary of past accounting performance. 

Thus, firms with excellent growth prospects tend to have high market‐to‐book ratios, and those 

with agency problems might have low ratios—and perhaps these considerations, rather than 

mispricing, drive investment and financing decisions. Dong, Hirshleifer, Richardson, and Teoh 

(2003) and Ang and Cheng (2005) discount analyst earnings forecasts to construct an arguably less 

problematic measure of fundamentals than book value. 

Another factor that limits this approach is that a precise ex ante measure of mispricing 

would represent a profitable trading rule. There must be limits to arbitrage that prevent rational 

investors from fully exploiting such rules and trading away the information they contain about 

mispricing.  

Ex post misvaluation. A second option is to use the information in future returns. The idea 

is that if stock prices routinely decline after a corporate event, one might infer that they were 

inflated at the time of the event. However, as detailed in Fama (1998) and Mitchell and Stafford 

(2000), this approach is also subject to critique.  

The most basic critique is the joint hypothesis problem: a predictable “abnormal” return 

might mean there was misvaluation ex ante, or simply that the definition of “normal” expected 

return (e.g., CAPM) is wrong. Perhaps the corporate event systematically coincides with changes in 

risk, and hence the return required in an efficient capital market. Another simple but important 

critique regards economic significance. Market value‐weighting or focusing on NYSE/AMEX firms 

may reduce abnormal returns or cause them to disappear altogether. 

Page 24: Behavioral Corporate Finance: An Updated Survey · Behavioral corporate finance, and behavioral finance more broadly, received a boost from the spectacular rise and fall of Internet

  20 

There are also statistical issues. For instance, corporate events are often clustered in time 

and by industry—IPOs are an example considered in Brav (2000)—and thus abnormal returns may 

not be independent. Barber and Lyon (1997) and Lyon, Barber, and Tsai (1999) show that 

inference with buy‐and‐hold returns (for each event) is challenging. Calendar‐time portfolios, 

which consist of an equal‐ or value‐weighted average of all firms making a given decision, have 

fewer problems here, but the changing composition of these portfolios adds another complication 

to standard tests. Loughran and Ritter (2000) also argue that such an approach is a less powerful 

test of mispricing, since the clustered events have the worst subsequent performance. A final 

statistical problem is that many studies cover only a short sample period. Schultz (2003) shows that 

this can lead to a small sample bias if managers engage in “pseudo” market timing, making 

decisions in response to past rather than future price changes.  

Analyzing aggregate time series resolves some of these problems. Like the calendar time 

portfolios, time series returns are more independent. There are also established time‐series 

techniques, e.g. Stambaugh (1999), to deal with small‐sample biases. Nonetheless, the joint 

hypothesis problem remains, since rationally required returns may vary over time. 

But even when these econometric issues can be solved, interpretational issues may remain. 

For instance, suppose investors have a tendency to overprice firms that have genuinely good 

growth opportunities. If so, even investment that is followed by low returns need not be ex ante 

inefficient. Investment may have responded to omitted measures of investment opportunities, not 

to the misvaluation itself.  

There are a variety of ways to improve the identification of a channel that connects capital 

market mispricing to corporate finance. Baker (2009) outlines an approach based on instrumenting 

for mispricing with investor tastes or other shocks to the supply of capital, and approaches 

involving the interaction of measures of valuation or mispricing with limits to arbitrage or 

Page 25: Behavioral Corporate Finance: An Updated Survey · Behavioral corporate finance, and behavioral finance more broadly, received a boost from the spectacular rise and fall of Internet

  21 

corporate incentives to time the market. Of course, even in these approaches using interaction 

terms, one still has to proxy for mispricing with an ex ante or ex post method. To the extent that the 

hypothesized cross‐sectional pattern appears strongly in the data, however, objections about the 

measure of mispricing lose some steam. 

Non­fundamental investor demand. The first approach is to identify supply effects with 

shifts in investor demand. The idea is to find empirical measures that are correlated with sentiment 

or the supply of capital but not with fundamentals. This is simple enough to write, but hard to 

implement. If it were possible to identify mispricing so clearly, such mispricing might not arise in 

the first place. Some examples are measures of investor inertia (Baker, Coval, and Stein (2007)), 

inattention (DellaVigna and Pollet (2009)), local demand (Becker, Ivkovic, and Weisbenner (2011)), 

overconfidence (combined with short sales constraints in Gilchrist, Himmelberg, and Huberman 

(2005)), or index additions (Massa, Peyer, and Tong (2005)). More broadly, shocks to the capital of 

intermediaries, while not necessarily behavioral, can be used to assess the impact of capital market 

inefficiency on corporate finance. This is too large a literature to survey here. This approach comes 

down to replacing a direct measure of valuation with an instrument for investor demand. 

Cross­sectional interactions: Limits to arbitrage. In situations where trading on 

mispricing is limited by short‐sales constraints, transaction costs, margin requirements, regulation, 

and fundamental risk, prices are likely to be further from fundamental value, making the impact of 

capital market inefficiencies on corporate finance more likely. For example, Baker, Foley, and 

Wurgler (2009) argue that the limits on arbitrage are more severe in some countries than others, 

leading to a differential effect of valuations of FDI. Lamont and Stein (2006) and Greenwood (2007) 

make similar arguments about relative efficiency the impact on stock issuance and mergers and 

acquisitions, and stock splits in Japan, respectively. This approach comes down to identifying 

market conditions where mispricing will have the strongest effect. 

Page 26: Behavioral Corporate Finance: An Updated Survey · Behavioral corporate finance, and behavioral finance more broadly, received a boost from the spectacular rise and fall of Internet

  22 

Cross­sectional interactions: Corporate opportunism. The effect of capital market 

inefficiencies on corporate finance should be most pronounced among those firms exhibiting the 

means and the incentive to be opportunistic. In this spirit, Baker, Stein, and Wurgler (2003) 

consider the prediction that if fe is positive, mispricing should be more relevant for financially 

constrained firms. More generally, managerial horizons or the fundamental costs of catering to 

sentiment may vary across firms in a measurable way. For example, Bergstresser and Phillipon 

(2006) show that earnings management is more pronounced when managers are compensated 

with stock and options. Gaspar, Massa, and Matos (2005) argue that managers inherit their 

investors’ incentives, which may not be chosen optimally to match firm fundamentals. This 

approach comes down to identifying firms where mispricing will have the strongest effect. 

2.4.  Investment policy 

Of paramount importance are the real consequences of market inefficiency. It is one thing to 

say that investor irrationality has an impact on capital market prices, or even financing policy, 

which leads to transfers of wealth among investors. It is another to say that mispricing leads to 

underinvestment, overinvestment, or the general misallocation of capital and deadweight losses for 

the economy as a whole. In this subsection we review research on how market inefficiency affects 

real investment, mergers and acquisitions, and diversification. 

2.4.1.   Real investment 

In the market timing and catering framework, mispricing influences real investment in two 

ways. First, investment may itself be a characteristic that is subject to mispricing (this happens 

when K is greater than zero above). Investors may overestimate the value of investment in 

particular technologies, for example. Second, a financially constrained firm (this can happen when 

feK is greater than zero above) may be forced to pass up fundamentally valuable investment 

opportunities if it is undervalued. 

Page 27: Behavioral Corporate Finance: An Updated Survey · Behavioral corporate finance, and behavioral finance more broadly, received a boost from the spectacular rise and fall of Internet

  23 

Most research has looked at the first type of effect. Of course, anecdotal evidence of this 

effect comes from bubble episodes; it was with the late 1920s bubble fresh in mind that Keynes 

(1936) argued that short‐term investor sentiment is, at least in some eras, a major or dominant 

determinant of investment. More recent US stock market episodes generally viewed as bubbles 

include the electronics boom in 1959‐62, growth stocks in 1967‐68, the “nifty fifty” in the early 

1970s, gambling stocks in 1977‐78, natural resources, high tech, and biotechnology stocks in the 

1980s, and the Internet in the late 1990s; see Malkiel (1990) for an anecdotal review of some of 

these earlier bubbles, and Ofek and Richardson (2003) on the Internet. See Kindleberger (2000) for 

an attempt to draw general lessons from bubbles and crashes over several hundred years, and for 

anecdotal remarks on their sometimes‐dramatic real consequences. 

An early wave of studies in this area tested whether investment is sensitive to stock prices 

over and above direct measures of the marginal product of capital, such as cash flow or profitability. 

If it is not, they reasoned, then the univariate link between investment and stock valuations likely 

just reflects the standard, efficient‐markets Q channel. This approach did not lead to a clear 

conclusion, however. For example, Barro (1990) argues for a strong independent effect of stock 

prices, while Morck, Shleifer, and Vishny (1990b) and Blanchard, Rhee, and Summers (1993) 

conclude that the incremental effect is weak.  

The more recent wave of studies takes a different tack. Rather than controlling for 

fundamentals and looking for a residual effect of stock prices, they try to proxy for the mispricing 

component of stock prices and examine whether it affects investment. In this spirit, Chirinko and 

Schaller (2001, 2004), Panageas (2003), Polk and Sapienza (2009), Gilchrist, Himmelberg, and 

Huberman (2005), Massa, Peyer, and Tong (2005), and Schaller (2011) all find evidence that 

investment is sensitive to proxies for mispricing. Of course, the generic concern is that the 

mispricing proxies are still just picking up fundamentals. To refute this, Polk and Sapienza as well 

Page 28: Behavioral Corporate Finance: An Updated Survey · Behavioral corporate finance, and behavioral finance more broadly, received a boost from the spectacular rise and fall of Internet

  24 

as Massa et al., for example, consider the finer prediction that investment should be more sensitive 

to short‐term mispricing when managerial horizons are shorter. Polk and Sapienza find that 

investment is indeed more sensitive to mispricing proxies when share turnover is higher, i.e., where 

the average shareholder’s horizon is shorter; the Massa et al. test is similar. 

The second type of mispricing‐driven investment is tested in Baker, Stein, and Wurgler 

(2003). Stein (1996) predicts that investment will be most sensitive to mispricing in equity‐

dependent firms, i.e. firms that have no option but to issue equity to finance their marginal 

investment, because long‐horizon managers of undervalued firms would rather underinvest than 

issue undervalued shares. Using several proxies for equity dependence and mispricing, Baker et al. 

confirm the prediction. 

Overall, there is some evidence that some portion of the effect of stock prices on investment 

is a response to mispricing, but key questions remain. The actual magnitude of the effect of 

mispricing has not been pinned down, even roughly. The efficiency implications are also unclear. 

Titman, Wei, and Xie (2004) and Polk and Sapienza (2009) find that high investment is associated 

with lower future stock returns in the cross section, and Lamont (2000) finds a similar result for 

planned investment in the time series. However, sentiment and fundamentals seem likely to be 

correlated, and so, as mentioned previously, even investment followed by low returns may not be 

ex ante inefficient.11 Even granting an empirical link between overpricing and investment, it is hard 

to determine the extent to which managers are rationally fanning the flames of overvaluation, as in 

catering, or are simply just as overoptimistic as their investors. We shall return to the effects of 

managerial optimism.  

                                                             

11 As an example of this complication, Campello and Graham (2007) find that financially strapped non‐tech firms issued equity during the Internet bubble and used it to invest. The unconstrained non‐tech firms did not show this pattern. This suggest that bubbles driven by one category can have positive spillover effects on relatively unrelated firms.  

Page 29: Behavioral Corporate Finance: An Updated Survey · Behavioral corporate finance, and behavioral finance more broadly, received a boost from the spectacular rise and fall of Internet

  25 

2.4.2.   Mergers and acquisitions 

Shleifer and Vishny (2003) propose a market‐timing model of acquisitions. They assume 

that acquirers are overvalued, and the motive for acquisitions is not to gain synergies, but to 

preserve some of their temporary overvaluation for long‐run shareholders. Specifically, by 

acquiring less‐overvalued targets with overpriced stock (or, less interestingly, undervalued targets 

with cash), overvalued acquirers can cushion the fall for their shareholders by leaving them with 

more hard assets per share. Or, if the deal’s value proposition caters to a perceived synergy that 

causes the combined entity to be overvalued, as might have happened in the late 1960s 

conglomerates wave (see below), then the acquirer can still gain a long‐run cushion effect, while 

offering a larger premium to the target.  

The market timing approach to mergers helps to unify a number of stylized facts. The 

defensive motive for the acquisition, and the idea that acquisitions are further facilitated when 

catering gains are available, help to explain the time‐series link between merger volume and stock 

prices, e.g., Golbe and White (1988).12 The model also predicts that cash acquirers earn positive 

long‐run returns while stock acquirers earn negative long‐run returns, consistent with the findings 

of Loughran and Vijh (1997) and Rau and Vermaelen (1998). 

Recent papers have found further evidence for market timing‐motivated mergers. Dong, 

Hirshleifer, Richardson, and Teoh (2003) and Ang and Cheng (2005) find that market‐level 

mispricing proxies and merger volume are positively correlated, and (within this) that acquirers 

tend to be more overpriced than targets.13 They also find that offers for undervalued targets are 

                                                             

12 See Rhodes‐Kropf and Viswanathan (2004) for a somewhat different misvaluation‐based explanation of this link.  

13 A related prediction of the Shleifer‐Vishny framework is that an overvalued acquirer creates value for long‐term shareholders by acquiring a fairly valued or simply less overvalued target. Savor and Lu (2009) tests this proposition by comparing the returns of successful acquirers to those that fail for exogenous reasons, 

Page 30: Behavioral Corporate Finance: An Updated Survey · Behavioral corporate finance, and behavioral finance more broadly, received a boost from the spectacular rise and fall of Internet

  26 

more likely to be hostile, and that overpriced acquirers pay higher takeover premia. Rhodes‐Kropf, 

Robinson, and Viswanathan (2005) also link valuations and merger activity. Bouwman, Fuller, and 

Nain (2003) find evidence suggestive of a short‐term catering effect. In high‐valuation periods, 

investors welcome acquisition announcements, yet the subsequent returns of mergers made in 

those periods are the worst. Baker, Foley, and Wurgler (2009) find that foreign direct investment 

(FDI), which is often cross‐border acquisitions, increase with the current aggregate market‐to‐book 

ratio of the source country stock market and decrease with subsequent returns on that market. All 

of these patterns are consistent with overvaluation‐driven merger activity. 

An unresolved question in the Shleifer‐Vishny framework is why managers would prefer a 

stock‐for‐stock merger to an equity issue if the market timing gains are similar. One explanation is 

that a merger more effectively hides the underlying market timing motive from investors, because 

the equity issue and investment decision are bundled. Baker, Coval, and Stein (2007) consider 

another mechanism that can also help explain a generic preference for equity issues via merger.14 

The first ingredient is that the acquiring firm faces a downward sloping demand curve for its 

shares. The second ingredient is that some investors follow the path of least resistance, passively 

accepting the acquirer’s shares as consideration even when they would not have actively 

participated in an equity issue. With these two assumptions, the price impact of a stock‐financed 

merger can be much smaller than the price impact of an SEO. Empirically, inertia is a major feature 

in institutional and especially individual holdings data that is associated with smaller merger 

announcement effects. 

                                                                                                                                                                                                    

such as a regulatory intervention. Successful acquirers perform poorly, as in Loughran and Vijh (1997), but unsuccessful acquirers perform even worse.  

14 For example, in the case of S&P 100 firms over 1999‐2001, Fama and French (2005) find that the amount of equity raised in mergers is roughly 40 times that raised in SEOs.  

Page 31: Behavioral Corporate Finance: An Updated Survey · Behavioral corporate finance, and behavioral finance more broadly, received a boost from the spectacular rise and fall of Internet

  27 

2.4.3.   Diversification and focus 

Standard explanations for entering unrelated lines of business include agency problems or 

synergies, e.g., internal capital markets and tax shields. Likewise, moves toward greater focus are 

often interpreted as triumphs of governance. While our main task is to survey the existing 

literature, the topics of diversification and focus have yet to be considered from a perspective 

where investors are less than fully rational. So, we take a short detour here. We ask whether the 

evidence at hand is consistent with the view that the late‐1960s conglomerate wave, which led to 

conglomerates so complex they were still being divested or busted up decades later, was in part 

driven by efforts to cater to a temporary investor appetite for conglomerates. 

Investor demand for conglomerates does appear to have reached a peak in 1968. 

Ravenscraft and Scherer (1987, p. 40) find that the average return on 13 leading conglomerates 

was 385% from July 1965 to June 1968, while the S&P 425 gained only 34%. Diversifying 

acquisitions were being greeted with a positive announcement effect, while other acquisitions were 

penalized (Matsusaka (1993)). Klein (2001) finds a “diversification premium” of 36% from 1966‐68 

in a sample of 36 conglomerates. Perhaps responding to these valuation incentives, conglomerate 

mergers accelerated in 1967 and peaked in 1968 (Ravenscraft and Scherer, pp. 24, 161, 218).  

Conglomerate valuations started to fall in mid‐1968. Between July 1968 and June 1970, the 

sample followed by Ravenscraft and Scherer lost 68%, three times more than the S&P 425. 

Announcement effects also suggest a switch in investor appetites: diversification announcements 

were greeted with a flat reaction in the mid‐ to late‐1970s and a negative reaction by the 1980s 

(Morck, Shleifer, and Vishny (1990a)). Klein finds that the diversification premium turned into a 

discount of 1% in 1969‐71 and 17% by 1972‐74, and a discount seems to have remained through 

the 1980s (Lang and Stulz (1994), Berger and Ofek (1995)). Again, possibly in response to this shift 

in catering incentives, unrelated segments began to be divested, starting a long trend toward focus 

Page 32: Behavioral Corporate Finance: An Updated Survey · Behavioral corporate finance, and behavioral finance more broadly, received a boost from the spectacular rise and fall of Internet

  28 

(Porter (1987), Kaplan and Weisbach (1992)).15 Overall, while systematic evidence is lacking, the 

drivers of the diversification and subsequent re‐focus wave could be related to catering.  

2.5.   Financial policy 

The simple theoretical framework suggests that long‐horizon managers may reduce the 

overall cost of capital paid by their ongoing investors by issuing overpriced securities and 

repurchasing underpriced securities. Next, we survey the evidence on the extent to which market 

timing affects equity issues, repurchases, debt issues, cross‐border issues, financial intermediation 

(with thoughts on the recent financial crisis), and capital structure. 

2.5.1.   Equity issues 

Several lines of evidence suggest that overvaluation is a motive for equity issuance. Most 

simply, in the Graham and Harvey (2001) anonymous survey of CFOs of public corporations, two‐

thirds state that “the amount by which our stock is undervalued or overvalued was an important or 

very important consideration” in issuing equity (p. 216). Several other questions in the survey also 

ask about the role of stock prices. Overall, stock prices are viewed as more important than nine out 

of ten factors considered in the decision to issue common equity, and the most important of five 

factors in the decision to issue convertible debt.  

Empirically, equity issuance is positively associated with plausible ex ante indicators of 

overvaluation. Pagano, Panetta, and Zingales (1998) examine the determinants of Italian private 

firms’ decisions to undertake an IPO between 1982 and 1992, and find that the most important is 

the market‐to‐book ratio of seasoned firms in the same industry. Lerner (1994) finds that IPO 

                                                             

15 In a case study of the diversification and subsequent refocus of General Mills, Donaldson (1990) writes that the company spent some effort “to verify the dominant trends in investor perceptions of corporate efficiency, as seen in the company study of the impact of excessive diversification on the trend of price‐earnings multiples in the 1970s” (p. 140).  

Page 33: Behavioral Corporate Finance: An Updated Survey · Behavioral corporate finance, and behavioral finance more broadly, received a boost from the spectacular rise and fall of Internet

  29 

volume in the biotech sector is highly correlated with biotech stock indexes. Loughran, Ritter, and 

Rydqvist (1994) find that aggregate IPO volume and stock market valuations are highly correlated 

in most major stock markets around the world. Similarly, Marsh (1982) examines the choice 

between (seasoned) equity and long‐term debt by UK quoted firms between 1959 and 1974, and 

finds that recent stock price appreciation tilts firms toward equity issuance. In US data, Jung, Kim, 

and Stulz (1996), Hovakimian, Opler, and Titman (2001), and Erel, Julio, Kim, and Weisbach (2010) 

also find a strong relationship between stock prices and seasoned equity issuance.  

There are many non‐mispricing reasons why equity issuance and market valuations should 

be positively correlated, of course. More specific evidence for equity market timing comes from the 

pattern that new issues earn low subsequent returns. In one of the earliest modern tests of market 

efficiency, Stigler (1964) tried to measure the effectiveness of the S.E.C. by comparing the ex post 

returns of new equity issues (lumping together both initial and seasoned) from 1923‐28 with those 

from 1949‐55. If the S.E.C. improved the pool of issuers, he reasoned, then the returns to issuers in 

the latter period should be higher. But he found that issuers in both periods performed about 

equally poorly relative to a market index. Five years out, the average issuer in the pre‐S.E.C. era 

lagged the market by 41%, while the average underperformance in the later period was 30%.  

Other sample periods show similar results. Ritter (1991) examines a sample of IPOs, Spiess 

and Affleck‐Graves (1995) examine SEOs, and Loughran and Ritter (1995) examine both.16 And, 

Ritter (2003) updates these and several other empirical studies of corporate financing activities. 

The last paper’s sample includes 7,437 IPOs and 7,760 SEOs between 1970 and 1990. Five years 

out, the average IPO earns lower returns than a size‐matched control firm by 30%, and the average 

SEO underperforms that benchmark by 29%. Gompers and Lerner (2003) fill in the gap between 

                                                             

16 Updated data on the long‐run returns of IPOs is available on Jay Ritter’s website at http://bear.warrington.ufl.edu/ritter/ipodata.htm.  

Page 34: Behavioral Corporate Finance: An Updated Survey · Behavioral corporate finance, and behavioral finance more broadly, received a boost from the spectacular rise and fall of Internet

  30 

the samples of Stigler (1964) and Loughran and Ritter (1995). Their sample of 3,661 IPOs between 

1935 and 1972 shows average five‐year buy‐and‐hold returns that underperform the value‐

weighted market index by 21% to 35%.17 Thus, a series of large and non‐overlapping samples 

suggests that, on average, US equity issues underperform the market somewhere in the ballpark of 

20‐40% over five years.  

In a test that speaks especially closely to opportunistic market timing of equity sales to new 

investors, Burch, Christie, and Nanda (2004) examine the subsequent performance of seasoned 

equity issued via rights offers, which are targeted to a firm’s ongoing shareholders, and firm 

commitment offers, which are targeted to new shareholders. In their 1933 to 1949 sample, a period 

in which rights offers were more common, they find underperformance concentrated entirely in the 

latter group. This fits the framework above, which emphasizes the opportunistic timing of equity 

sales to new investors. 

Much evidence suggests that investor sentiment varies over time in its strength and nature. 

For example, stock market bubbles can grow and pop within certain industries. Greenwood and 

Hanson (2011) exploit this observation. They find that net equity issuance by firms with different 

characteristics—size, share price, distress status, payout policy, industry, and profitability—helps 

to predict returns on portfolios defined on those characteristics. Their paper is also an interesting 

contribution to behavioral asset pricing and shows the value of a unified perspective. That is, the 

paper suggests how the misvaluation of firm characteristics at any given point in time, an otherwise 

difficult concept to measure, is betrayed by the financing activity and market timing motives of 

firms. We will see more results of this sort in the catering section. 

                                                             

17 Gompers and Lerner also confirm what Brav and Gompers (1997) found in a later sample: while IPOs have low absolute returns, and low returns relative to market indexes, they often do not do worse than stocks of similar size and book‐to‐market ratio. One interpretation is that securities with similar characteristics, whether or not they are IPOs, tend to be similarly priced (and mispriced) at a given point in time. 

Page 35: Behavioral Corporate Finance: An Updated Survey · Behavioral corporate finance, and behavioral finance more broadly, received a boost from the spectacular rise and fall of Internet

  31 

If equity issues cluster when the market as a whole is overvalued, the net gains to equity 

market timing may be even larger than the underperformance studies suggest. Baker and Wurgler 

(2000) examine whether equity issuance, relative to total equity and debt issuance, predicts 

aggregate market returns between 1927 and 1999. They find that when the equity share was in its 

top historical quartile, the average value‐weighted market return over the next year was negative 

6%, or 15% below the average market return. Henderson, Jegadeesh, and Weisbach (2006) find a 

similar relationship in several international markets over the period 1990 to 2001. In 12 out of the 

13 markets they examine, average market returns are higher after a below‐median equity share 

year than after an above‐median equity share year.18 

The equity market timing studies continue to be hotly debated. Some authors highlight the 

usual joint hypothesis problem, implicitly proposing that IPOs and SEOs deliver low returns 

because they are actually far less risky (and priced accordingly by investors). This notion strikes us 

as fanciful, but for more on this perspective, see Eckbo, Masulis, and Norli (2000),and Eckbo and 

Norli (2004). On a statistical point, Schultz (2003) highlights a small‐sample “pseudo market 

timing” bias that can lead to exaggerated impressions of underperformance when abnormal 

performance is calculated in “event time.” The empirical relevance of this bias is unclear. Schultz 

(2003, 2004) argues that it may be significant, while Ang, Gu, and Hochberg (2007), Dahlquist and 

de Jong (2004), and Viswanathan and Wei (2008) argue that it is minor.19 The key issue concerns 

                                                             

18 Note that these aggregate predictability results should probably not be interpreted as evidence that “managers can time the aggregate market.” A more plausible explanation is that broad waves of investor sentiment lead many firms to be mispriced in the same direction at the same time. Then, the average financing decision will contain information about the average (i.e., market‐level) mispricing, even though individual managers are perceiving and responding only to their own firm’s mispricing. 

19 Butler, Grullon, and Weston (2005) take Schultz’s idea to the time‐series and argue that the equity share’s predictive power is due to an aggregate version of the pseudo market timing bias. Baker, Taliaferro, and Wurgler (2006) reply that the tests in Butler et al. have little actual relevance to the bias and that standard econometric techniques show that small‐sample bias can account for only one percent of the equity share’s actual predictive coefficient.  

Page 36: Behavioral Corporate Finance: An Updated Survey · Behavioral corporate finance, and behavioral finance more broadly, received a boost from the spectacular rise and fall of Internet

  32 

the variance in the number of security issues over time. Schultz assumes a nonstationary process 

for this time series. This means that the number of security issues can explode or collapse to zero 

for prolonged periods of time, and his simulated variance of equity issuance exceeds the actual 

experience in the U.S. 

In any case, the returns studies, having by nature low power, should not be considered in 

isolation. Survey evidence was mentioned above. Other relevant results include Teoh, Welch, and 

Wong (1998a,b), who find that the equity issuers who manage earnings most aggressively have the 

worst post‐issue returns. Jain and Kini (1994), Mikkelson, Partch, and Shah (1997), and Pagano et 

al. (1998) find that profitability deteriorates rapidly following the initial offering, and Loughran and 

Ritter (1997) document a similar pattern with seasoned issues. Insider selling also coincides with 

seasoned offerings, Jenter (2005) finds. In a roundabout but novel approach, DellaVigna and Pollet 

(2011) hypothesize that managers but not investors recognize the effect of demographic shifts on 

stock prices in the next five to ten years. Under a market timing policy, managers will wait for those 

shifts to push up (down) prices to issue (repurchase) equity; perhaps surprisingly, they find 

evidence for such an effect.    

Market timing can help resolve a puzzle of how or why issuers are able to raise outside 

equity when potential agency costs are high. In the traditional view of Jensen and Meckling (1976), 

existing owners bear future agency costs up front when they raise new equity, potentially 

rendering outside equity prohibitively costly. This assumes of course that outside investors are 

rationally computing these costs. Chernenko, Greenwood, and Foley (2010) find that Japanese firms 

with the highest agency costs appear to raise capital when perceptions of agency costs are low. After 

listing, their subsequent performance is very poor, as if investors periodically ignored potential 

agency problems. 

Page 37: Behavioral Corporate Finance: An Updated Survey · Behavioral corporate finance, and behavioral finance more broadly, received a boost from the spectacular rise and fall of Internet

  33 

Viewed as a whole, the evidence indicates that market timing and attempted market timing 

play a considerable role in equity issuance decisions. That said, DeAngelo, DeAngelo, and Stulz 

(2010) remind us that seasoned equity issuance that is not associated with mergers is still an 

infrequent event.   

2.5.2.   Repurchases 

Undervaluation is a very important motive for repurchases. Brav, Graham, Harvey, and 

Michaely (2005) survey 384 CFOs regarding payout policy, and “the most popular response for all 

the repurchase questions on the entire survey is that firms repurchase when their stock is a good 

value, relative to its true value: 86.6% of all firms agree” (p. 26). Anecdotally, repurchases cluster 

after unusual market crashes: Hong, Wang, and Yu (2008) highlight the repurchase waves that 

followed after crashes in October 1987 and September 11, 2001.  

At the firm level, repurchasers earn positive abnormal returns on average, suggesting that 

managers are on average successful in timing them. Ikenberry, Lakonishok, and Vermaelen (1995) 

study 1,239 open market repurchases announced between 1980 and 1990. Over the next four 

years, the average repurchaser earned 12% more than firms of similar size and book‐to‐market 

ratios. Ikenberry, Lakonishok, and Vermaelen (2000) find similar results in a sample of Canadian 

firms. Note that these returns are benchmark‐adjusted and therefore do not count any successful 

timing of repurchases from, for example, the rebound from the October 1987 crash.20  

  The evidence is that managers tend to issue equity before low returns, on average, and 

repurchase before higher returns. Without knowing just how the “rational” cost of equity varies 

over time, it is difficult to know how much this activity actually reduces the cost of equity for the 

average firm. However, suppose that rationally required returns are constant. By following 

                                                             

20 Baker and Wurgler (2000) also study the ability of net equity issuance to predict market returns.  

Page 38: Behavioral Corporate Finance: An Updated Survey · Behavioral corporate finance, and behavioral finance more broadly, received a boost from the spectacular rise and fall of Internet

  34 

aggregate capital inflows and outflows into corporate equities, and tracking the returns that follow 

these flows, Dichev (2004) reports that the average “dollar‐weighted” return is lower than the 

average buy‐and‐hold return by 1.3% per year for the NYSE/Amex, 5.3% for Nasdaq, and 1.5% (on 

average) for 19 stock markets around the world. Put differently, if NYSE/Amex firms had issued 

and repurchased randomly across time, then, holding the time series of realized returns fixed, they 

would have paid 1.3% per year more for the equity capital they employed.  

Of course, this reduction in the cost of equity capital is not evenly distributed in the cross 

section of firms. The composition of firms in position to repurchase, for example, varies over time, 

as shown by Greenwood and Hanson (2011), in accord with valuation. The static difference 

between Nasdaq and NYSE/Amex also gives a hint of this. For the many mature firms that rarely 

raise external equity, the gains may be negligible. For other firms that access the capital markets 

repeatedly through seasoned equity issues and stock‐financed mergers, the gains may be much 

larger.  

2.5.3.   Debt issues 

A few papers have examined debt market timing—raising debt when its cost is unusually 

low. Survey evidence offers support for market timing being a factor in debt issuance decisions. 

Graham and Harvey (2001) find that interest rates are the most cited factor in debt policy 

decisions: CFOs issue debt when they feel “rates are particularly low.” Expectations about the yield 

curve also appear to influence the maturity of new debt. Short‐term debt is preferred “when short‐

term rates are low compared to long‐term rates” and when “waiting for long‐term market interest 

rates to decline.” While the former statement would be consistent with the preference for a low 

interest rates to pump up earnings (Stein (1989)), the latter clearly indicates a skepticism in the 

textbook expectations hypothesis, which posits that the cost of debt is equal across maturities. At 

Page 39: Behavioral Corporate Finance: An Updated Survey · Behavioral corporate finance, and behavioral finance more broadly, received a boost from the spectacular rise and fall of Internet

  35 

the same time, CFOs do not confess to exploiting their private information about credit quality, 

instead highlighting general debt market conditions. 

On the empirical side, Marsh (1982), in his sample of UK firms, finds that the choice 

between debt and equity does appear to be swayed by the level of interest rates. Guedes and Opler 

(1996) examine and largely confirm the survey responses regarding the effect of the yield curve. In 

a sample of 7,369 US debt issues between 1982 and 1993, they find that maturity is strongly 

negatively related to the term spread (the difference between long‐ and short‐term bond yields), 

which fluctuated considerably during this period.  

Is there any evidence that debt market timing is successful? In aggregate data, Baker, 

Greenwood, and Wurgler (2003) examine the effect of debt market conditions on the maturity of 

debt issues and, perhaps more interestingly, connect the maturity of new issues to subsequent bond 

market returns. Specifically, in US Flow of Funds data between 1953 and 2000, the aggregate share 

of long‐term debt issues in total long‐ and short‐term debt issues is negatively related to the term 

spread, just as Guedes and Opler find with firm‐level data. Further, because the term spread is 

positively related to future excess bond returns—i.e. the difference in the returns of long‐term and 

short‐term bonds, or the realized relative cost of long‐ and short‐term debt—so is the long‐term 

share in debt issues. Perhaps simply by using a naïve rule of thumb, “issue short‐term debt when 

short‐term rates are low compared to long‐term rates,” managers may have timed their debt 

maturity decisions so as to reduce their overall cost of debt. Of course, such a conclusion is subject 

to the usual risk‐adjustment caveats. 

Greenwood, Hanson, and Stein (2008) go deeper into the effect of debt market efficiency on 

maturity structure, and while it falls within the market timing spirit it has the appealing feature that 

it does not require that firms have a debt market forecasting ability. Specifically, they argue that 

there are shocks to supply of bonds at different points in the yield curve, for example changes in the 

Page 40: Behavioral Corporate Finance: An Updated Survey · Behavioral corporate finance, and behavioral finance more broadly, received a boost from the spectacular rise and fall of Internet

  36 

maturity structure of government debt, that introduce corresponding mispricings along the yield 

curve. Anyone can observe these. Given limited arbitrage on the investor side, firms that are 

indifferent to their debt maturity (in this otherwise Modigliani‐Miller world) can supply debt at the 

mispriced term, limited only by their size.  

Unfortunately, the data on individual debt issues and their subsequent returns does not 

approach the level of detail of the IPO and SEO data. But one intriguing pattern that has been 

uncovered is that debt issues, much like equity issues, are followed by low equity returns. Spiess 

and Affleck‐Graves (1999) examine 392 straight debt issues and 400 convertible issues between 

1975 and 1989. The shares of straight debt issuers underperform a size‐ and book‐to‐market 

benchmark by an insignificant 14% over five years (the median underperformance is significant), 

while convertible issuers underperform by a significant 37%. There is also a suggestion that the 

riskiest firms may be timing their idiosyncratic credit quality, despite the survey answers on this 

point: the shares of unrated issuers have a median five‐year underperformance of 54%. If the 

equity did so poorly, the debt issues presumably also did poorly. In a much broader panel, 

Richardson and Sloan (2003) also find that net debt issuance is followed by low stock returns.  

  There are several potential explanations for this pattern. Certainly, equity overvaluation 

would be expected to lower the cost of debt directly, because credit risk models routinely include 

stock market capitalization as an input, so the relationship with subsequent stock returns may 

reflect debt market timing per se. Or, perhaps managerial and investor sentiment is correlated; 

managers may tend to be most optimistic precisely when capital is cheap, and thus raise and invest 

as much as they can from any source. This story combines investor and managerial irrationality and 

so does not fit neatly within the market timing framework, but may have some truth. A third 

possibility, outlined in Baker, Stein, and Wurgler (2003), is that equity overvaluation relaxes a 

Page 41: Behavioral Corporate Finance: An Updated Survey · Behavioral corporate finance, and behavioral finance more broadly, received a boost from the spectacular rise and fall of Internet

  37 

binding leverage constraint, creating debt capacity that subsequently gets used up. But debt is 

always correctly priced in this setting, so debt market timing per se is not possible. 

2.5.4.   Cross‐border issues 

  The study of dual‐listed shares by Froot and Dabora (1999) shows that even highly liquid 

markets such as the US and the UK can attach different prices to the same cash flow stream. This 

raises the possibility of timing across international markets. Along these lines, Graham and Harvey 

(2001) find that among US CFOs who have considered raising debt abroad, 44% implicitly 

dismissed covered interest parity in replying that lower foreign interest rates were an important 

consideration in their decision.21 

  In practice, most international stock and bond issues are made on the US and UK markets. 

Henderson, Jegadeesh, and Weisbach (2006) find that when total foreign issues in the US or the UK 

are high, relative to respective GDP, subsequent returns on those markets tend to be low, 

particularly in comparison to the returns on issuers’ own markets. In a similar vein, and consistent 

with the survey evidence mentioned above, foreign firms tend to issue more debt in the US and the 

UK when rates there are low relative to domestic rates. 

2.5.5.   Financial intermediation 

  Our focus is mostly on the financing decisions of firms, but financial intermediaries often 

play a critical role between firms and the ultimate investors. To the extent that capital market 

inefficiencies affect corporate finance, an interesting question is how intermediaries affect issuance 

and investment patterns and whether they play a stabilizing or destabilizing role. The role of 

financial intermediaries in behavioral corporate finance is an interesting question in its own right 

that deserves more research attention. We mention papers in the area of banking, private equity, 

                                                             

21 Almost all equity raised by US corporations is placed in domestic markets, so Graham and Harvey do not ask about the determinants of international stock issues.  

Page 42: Behavioral Corporate Finance: An Updated Survey · Behavioral corporate finance, and behavioral finance more broadly, received a boost from the spectacular rise and fall of Internet

  38 

and venture capital. These questions obviously loom large in light of the recent financial crisis, 

which we discuss next.  

Banks are not dissimilar to firms in that they have the same market timing motives to sell 

overvalued securities and buy back securities that are undervalued. Motivated by the crisis, Shleifer 

and Vishny (2010) model how financial intermediaries can take advantage of investor sentiment in 

this way through securitized lending—creating and selling overpriced assets. This creates a channel 

for banks to transmit sentiment‐driven mispricing into real effects. In their model, banks retain a 

fraction of their loans. After a haircut, the value of these loans determines how much they can 

borrow short‐term.  When loan values are high, borrowing to make more of them and expand the 

balance sheet and finance more real investment is so profitable that it is worth the risk of having to 

liquidate their holdings if and when prices fall below fundamentals. As Charles Prince, the CEO of 

Citigroup, famously said in July 2007, “When the music stops, in terms of liquidity, things will be 

complicated. But as long as the music is playing, you’ve got to get up and dance. We’re still dancing.” 

As a result, far from being in a position to buy underpriced loans and stabilize the market, or 

finance new investment, banks can deepen a crisis.  

Fang, Ivashina, and Lerner (2010) find evidence of opportunism in bank involvement in 

private equity. In particular, banks’ share of private equity transactions peaks when the private 

equity market is experiencing large inflows. Moreover, transactions done at market peaks are more 

likely to turn out poorly. A broader view of private equity is that it profits from the imperfect 

integration between credit and equity markets. Occasionally, borrowing to finance the purchase of 

public or private firms is cheap relative to the cost of equity capital, enticing the share of private 

equity in mergers and acquisitions to cheap. Because this is purely a time series view, and private 

equity has a short history, it is difficult to prove. However, Axelson, Jenkinson, Stromberg, and 

Weisbach (2010) provide corroborating evidence of a link between financing costs and deal pricing. 

Page 43: Behavioral Corporate Finance: An Updated Survey · Behavioral corporate finance, and behavioral finance more broadly, received a boost from the spectacular rise and fall of Internet

  39 

It has been suggested that intermediaries can cause financial market “dislocations” to 

propagate from one set of firms to another, affecting real activity. Townsend (2011) considers the 

case of venture capital, where information asymmetry can lead to the portfolio firm being locked 

into a relationship with one capital provider, e.g. as in Rajan (1992). He finds that after the Internet 

bubble burst, non‐tech firms had difficulty getting follow‐on funding if their venture capitalists had 

high tech exposure. The question is why venture capitalists do not respond by diversifying their 

portfolios or reserving capital for follow‐on offerings. This is in the same spirit as the Shleifer‐

Vishny model, where in this case the lure of reselling Internet firms to a frothy market is so 

profitable that it is worth the risk of being short of capital in the event of a collapse. 

The recent financial crisis has many different elements, from the decisions of individual 

borrowers to the ultimate purchasers of mortgage backed securities, and the involvement of 

numerous intermediaries, including mortgage brokers, mortgage banks, investment banks and 

other underwriters of mortgage‐backed and other collateralized debt obligations (CDOs), ratings 

agencies, bond insurers, and the government‐sponsored entities, Fannie Mae and Freddie Mac. It is 

no surprise that there is not a tidy behavioral, or rational, explanation to its causes or its ultimate 

real consequences. Barberis (2011) makes significant progress in this direction. We do not have 

room to fully survey the burgeoning literature on the crisis here.  

A behavioral view of the crisis starts with the observation that less than fully rational 

demand was the underpinning of twin bubbles in real estate and the debt contracts underlying real 

estate and other similar assets. There are a variety of explanations. For example, investors and 

ratings agencies neglected a rare but not zero probability bad state and overvalued quasi‐AAA 

securities in Gennaioli, Shleifer, and Vishny (2011). Real estate and credit instruments were 

difficult to short, so differences of opinion may have led to overvaluation. Or, most simply, investors 

extrapolated short histories of high real estate returns and low default probabilities. Greenwood 

Page 44: Behavioral Corporate Finance: An Updated Survey · Behavioral corporate finance, and behavioral finance more broadly, received a boost from the spectacular rise and fall of Internet

  40 

and Hanson (2010) find predictability in a much longer time series of returns on credit. A period of 

high returns on risky debt and loosened credit standards is predictably followed by lower returns. 

Institutions played a role, catering to investor demand for safe assets. Investment banks 

created seemingly low risk assets with pooling and tranching. This combined in some cases with 

bond insurance increased the supply for AAA securities. Coval and Stafford (2010) argue that 

ratings agencies focused on default probabilities, neglecting the price of risk for senior tranches of 

CDOs. This is a more subtle argument than the conflicts of interest of issuers paying the ratings 

agencies for an opinion that have been highlighted by politicians and the media. 

A defining feature of the financial crisis was that systemically important banks retained a 

significant exposure to all types of mortgage securities. There are a number of explanations. One is 

that they simply carried inventory of mortgages and were left with these securities on their balance 

sheets at the start of the financial crisis. Unlike Internet IPOs, CDOs required time and bank capital 

to assemble. A second explanation is that they intentionally took risks with limited bank capital, 

intentionally gambling on a positive outcome in the mortgage markets. This moral hazard view has 

shaped the debate in financial reform. A challenge to this view is that the leadership of Bear Stearns 

and Lehman Brothers who were in a position to change leverage had a lot at stake, and indeed lost 

much of their wealth in 2008. A third explanation is that there were agency problems within the 

firm and the structured finance groups with the most information about these markets did not 

share with management. A final explanation is that they were convinced by their own marketing or, 

relatedly, they were focused on short‐term performance and the high prices of mortgage securities 

that changed hands prior to the crisis. This belongs to the section on less than fully rational 

managers. Whether this was overconfidence, cognitive dissonance, or a larger sociological 

phenomenon is hard to pin down. 

Page 45: Behavioral Corporate Finance: An Updated Survey · Behavioral corporate finance, and behavioral finance more broadly, received a boost from the spectacular rise and fall of Internet

  41 

A few general observations are worth making about recent financial crises. Both the 

Internet crash and the financial crisis started with significant asset price bubbles, both also 

involved the active or at least complicit participation of financial intermediaries, but the financial 

crisis involved much more direct exposure within the banking system—and hence larger real 

consequences. Moreover, both seem to involve equal parts of agency problems within institutions 

and investor sentiment. 

2.5.6.   Capital structure 

As an accounting identity, a firm’s capital structure is the cumulative outcome of a long 

series of incremental financing decisions, each driven by the need to fund some investment project, 

consummate a merger, refinance or rebalance, or achieve some other purpose. To the extent that 

market timing is a determinant of any of these incremental financing decisions, then, it may help to 

explain the cross‐section of capital structure. In particular, if market timing‐motivated financing 

decisions are not quickly rebalanced away, low‐leverage firms will tend to be those that raised 

external finance when their stock prices were high, and hence those that tended to choose equity to 

finance past investments and mergers, and vice‐versa for high leverage firms.22  

Such a market timing theory of capital structure is outlined in Baker and Wurgler (2002). In 

an effort to capture the historical coincidence of market valuations and the demand for external 

finance in a single variable, they construct an “external finance weighted‐average” of a firm’s past 

market‐to‐book ratios. For example, a high value would mean that the firm raised the bulk of its 

external finance, equity or debt, when its market‐to‐book was high. If market timing has a 

persistent impact on capital structure, this variable will have a negative cross‐sectional relationship 

                                                             

22 Similarly, debt maturity structure could to some extent reflect the historical coincidence of debt‐raising needs and debt market conditions like the term spread.  

Page 46: Behavioral Corporate Finance: An Updated Survey · Behavioral corporate finance, and behavioral finance more broadly, received a boost from the spectacular rise and fall of Internet

  42 

to the debt‐to‐assets ratio, even in regressions that control for the current market‐to‐book ratio. In 

a broad Compustat sample from 1968 to 1999, a strong negative relationship is apparent.  

  This evidence has inspired debate. On one hand, Hovakimian (2006) argues that equity 

issues do not have persistent effects on capital structure, and that the explanatory power of the 

weighted average market‐to‐book arises because it contains information about growth 

opportunities, a likely determinant of target leverage, that is not captured in current market‐to‐

book. Leary and Roberts (2005), Kayhan and Titman (2004), Flannery and Rangan (2006) also 

argue that firms rebalance toward a target. Alti (2006) looks specifically at the time series variation 

in IPO leverage, finding that an initial and statistically significant response to hot issues markets is 

short‐lived.  

On the other hand, Huang and Ritter (2009) show that the tendency to fund a financing 

deficit with equity decreases when the cost of equity is low. Furthermore, Welch (2004) and Huang 

and Ritter (2009), like Fama and French (2002), argue that firms rebalance their capital structures 

much more slowly, so that shocks to capital structure are long lived. And, in any event, Chen and 

Zhao (2007) point out that mean reversion in leverage is not definitive evidence for a tradeoff 

theory. Leverage is a ratio, so shocks tend to cause mean reversion mechanically. In an analysis of 

the choice between equity and debt issues, which avoids this problem, Chen and Zhao (2005) find 

that deviation‐from‐target proxies have little explanatory power, while market‐to‐book and past 

stock returns are very important. 

Page 47: Behavioral Corporate Finance: An Updated Survey · Behavioral corporate finance, and behavioral finance more broadly, received a boost from the spectacular rise and fall of Internet

  43 

2.6.   Other corporate decisions 

In this subsection, we consider what the market timing and catering approach has to say 

about dividend policy, firm name changes, and earnings management.23 We also discuss work that 

looks at executive compensation from this perspective. 

2.6.1.   Dividends 

The catering idea has been applied to dividend policy. Long (1978) provides some early 

motivation for this application. He finds that shareholders of Citizens Utilities put different prices 

on its cash dividend share class than its stock dividend share class, even though the value of the 

shares’ payouts are equal by charter. In addition, this relative price fluctuates. The unique 

experiment suggests that investors may view cash dividends per se as a salient characteristic, and in 

turn raises the possibility of a catering motive for paying them. 

Baker and Wurgler (2004a) test a catering theory of dividends in aggregate US data 

between 1963 and 2000. They find that firms initiate dividends when the shares of existing payers 

are trading at a premium to those of nonpayers, and dividends are omitted when payers are at a 

discount. To measure the relative price of payers and nonpayers, they use an ex ante measure of 

mispricing they call the “dividend premium,” which is just the difference between the average 

market‐to‐book ratios of payers and nonpayers. They also use ex post returns, and find that when 

the rate of dividend initiation increases, the future stock returns of payers (as a portfolio) are lower 

than those of nonpayers. This is consistent with the idea that firms initiate dividends when existing 

payers are relatively overpriced. Li and Lie (2006) find similar results for dividend changes. 

                                                             

23 We put dividend policy in this section and repurchases in the financing section, because, unlike a repurchase, pro‐rata dividends do not change the ownership structure of the firm, and there is no market timing benefit or cost. For this reason, it fits more naturally with the category of corporate decisions that might influence the level of mispricing, but do not by themselves transfer value among investors. 

Page 48: Behavioral Corporate Finance: An Updated Survey · Behavioral corporate finance, and behavioral finance more broadly, received a boost from the spectacular rise and fall of Internet

  44 

Time‐varying catering incentives shed much light on the “disappearance” of dividends. 

Fama and French (2001) document that the percentage of Compustat firms that pay dividends 

declines from 67% in 1978 to 21% in 1999, and that only a part of this is due to the compositional 

shift towards small, unprofitable, growth firms which are generally less likely to pay dividends. 

Baker and Wurgler (2004b) document that the dividend premium switched sign from positive to 

negative in 1978 and has remained negative through 1999, suggesting that dividends may have 

been disappearing in part because of the consistently lower valuations put on payers over this 

period. An analysis of earlier 1963‐77 data also lends support to this idea. Dividends “appeared,” 

“disappeared,” and then “reappeared” in this period, and each shift roughly lines up with a flip in 

the sign of the dividend premium. In UK data, Ferris, Sen, and Yui (2006) find that dividends have 

been disappearing during the late 1990s, and that a dividend premium variable formed using UK 

stocks lines up with this pattern.  

Supposing that dividend supply does respond to catering incentives, why does investor 

demand for payers vary over time in the first place? One possibility is that “dividend clienteles” 

vary over time, for example with tax code changes. However, in US data, the dividend premium is 

unrelated to the tax disadvantage of dividend income, as is the rate of dividend initiation. Shefrin 

and Statman (1984) develop explanations for why investors prefer dividends based on self‐control 

problems, prospect theory, mental accounting, and regret aversion. Perhaps these elements vary 

over time. Baker and Wurgler (2004a) argue that the dividend premium reflects sentiment for 

“risky” nonpaying growth firms versus “safe” dividend payers, since it falls in growth stock bubbles 

and rises in crashes; Fuller and Goldstein (2011) show explicitly that payers outperform in market 

downturns. Anecdotal evidence suggests that some investors flock to the perceived safety of 

dividends in gloomy periods, and bid up payers’ prices, at least in relative terms, in the process.  

Page 49: Behavioral Corporate Finance: An Updated Survey · Behavioral corporate finance, and behavioral finance more broadly, received a boost from the spectacular rise and fall of Internet

  45 

There are limitations to a catering theory of dividends. For one, it is a descriptive theory of 

whether firms pay dividends at all, not how much—in US data, at least, the dividend premium does 

not explain aggregate fluctuations in the level of dividends. DeAngelo, DeAngelo, and Skinner 

(2004) report that the aggregate dollar value of dividends has increased in real terms, as dividends 

have become concentrated in a smaller faction of traded firms. Also, the theory works better for 

explaining initiations than omissions, and it has little to say about the strong persistence in 

dividend policy. Catering, like agency or asymmetric information or taxes, is best viewed as one 

element in an overall theory of dividend policy. As we will see later, it is not even the only approach 

to dividends that behavioral corporate finance offers.   

2.6.2.   Earnings management 

  The quarterly net income figure that managers report to shareholders differs from actual 

economic cash flows by various non‐cash accruals, some discretionary. This becomes interesting 

when, as documented in the survey by Graham, Harvey, and Rajgopal (2005), CFOs believe that 

investors care more about earnings per share than cash flows.  

  Indeed, certain patterns in reported earnings numbers are clearly shaped by catering 

concerns. Most prominent among these are the reference points documented by Degeorge, Patel, 

and Zeckhauser (1999). Earnings are managed to exceed three salient thresholds. In order of 

importance, these are positive earnings, past reported earnings, and analysts’ expectations. 

Interestingly, the shape of the earnings distributions show that the threshold is generally met from 

below: firms near the thresholds stretch to meet them, not treating them as lower bounds and 

shifting earnings to the future.24 Carslaw (1988) and Bernard (1989) find that reported earnings 

and earnings per share cluster at salient round numbers, such as multiples of five or ten cents. 

                                                             

24 In the behavioral signaling section of the paper, we discuss a more dynamic model with both features. 

Page 50: Behavioral Corporate Finance: An Updated Survey · Behavioral corporate finance, and behavioral finance more broadly, received a boost from the spectacular rise and fall of Internet

  46 

These patterns do not hold for negative earnings, however; apparently, managers do whatever they 

can to distract attention from bad results.  

  These patterns have a flavor of catering to shareholder loss aversion relative to salient 

earnings reference points. At the same time, there are non‐behavioral contributors to these 

patterns. First, earnings management can be a Nash equilibrium result (Stein (1989)) under 

asymmetric information. Second, managerial bonuses or debt contracts may be conditional on 

earnings performance relative to simple benchmarks. Of course, the use of such contracts begs the 

question why shareholders and financiers should care about salient benchmarks over continuous 

measures of performance in the first place.  

Consistent with catering, managers with “short horizons” are especially likely to manage 

earnings. Bergstresser and Philippon (2006) find that accruals management is greater in companies 

whose CEO’s compensation, via stock and options holdings, is sensitive to current share prices. 

Sloan (1996) finds that firms with high accruals earn low subsequent returns, which suggests that 

earnings management may be successful in boosting share price, or at least in sustaining 

overvaluation. Consistent with the view that managers use earnings management to fool investors 

and issue overvalued equity, Teoh, Welch, and Wong (1998a,b) find that initial and seasoned equity 

issuer underperformance is greatest for firms that most aggressively manage pre‐issue earnings.  

An important question is whether earnings management has significant consequences for 

investment. Graham, Harvey, and Rajgopal (2005) present CFOs with hypothetical scenarios and 

find that 41% of them would be willing to pass up a positive‐NPV project just to meet the analyst 

consensus EPS estimate. Direct evidence of this type of value loss is difficult to document, but 

Jensen (2005) presents several anecdotes, and suggestive empirical studies include Teoh et al. 

(1998a,b), Erickson and Wang (1999), Bergstresser, Desai, and Rauh (2006), and McNichols and 

Page 51: Behavioral Corporate Finance: An Updated Survey · Behavioral corporate finance, and behavioral finance more broadly, received a boost from the spectacular rise and fall of Internet

  47 

Stubben (2008). One provocative finding is that earnings management activity increases prior to 

stock acquisitions.  

2.6.3.   Firm names 

Name changes provide some of the simplest and most colorful examples of catering. In 

frictionless and efficient markets, of course, firm names are as irrelevant as dividends. But there is 

at least a modest fundamental cost of changing names, and perhaps through a name change a firm 

can create a salient association with a temporarily overpriced category of stocks.  

Evidence of a catering motive for corporate names is most prominent in bubbles. In the 

1959‐62 era which Malkiel (1990) refers to as the “tronics boom,” firms “often included some 

garbled version of the word ‘electronics’ in their title even if the companies had nothing to do with 

the electronics industry” (p. 54). Systematic evidence has been assembled for the Internet bubble. 

Cooper, Dimitrov, and Rau (2001) find that 147 (generally small) firms changed to “dotcom” names 

between June 1998 and July 1999, as Internet valuations were rapidly rising. Catering to Internet 

sentiment did seem to deliver a short‐term price boost: Cooper et al. report a remarkably large 

average announcement effect of 74% for their main sample, and an even larger effect for the subset 

that had little true involvement with the Internet.  

Interestingly, Cooper et al. (2005) document that names were later used to dissociate 

companies from the Internet sector when prices crashed. Between August 2000 and September 

2001, firms that dropped their dotcom name saw a positive announcement effect of around 70%. 

The effect was almost as large for firms that dropped the dotcom name but kept an Internet 

business focus, and for the double dippers which dropped the name they had newly adopted just a 

few years earlier.  

Mutual fund companies also appear to be aware of the power that names have on investor 

demand. Cooper, Gulen, and Rau (2005) find that fund names shift away from styles that experience 

Page 52: Behavioral Corporate Finance: An Updated Survey · Behavioral corporate finance, and behavioral finance more broadly, received a boost from the spectacular rise and fall of Internet

  48 

low returns and toward those with high returns. The authors find that name changes do not predict 

fund performance, yet inflows increase dramatically, even for cosmetic name changers whose 

underlying investment style remains constant. Presumably, then, the name change decision is 

driven in part by the desire to attract fund inflows and thus increase the fund’s fee income. Indeed, 

Cooper et al. find that the inflow effect increases when money is spent to advertise the “new” styles.  

2.6.4.  Nominal share prices 

The average share price has centered around $25 since the Depression, as noted by Dyl and 

Elliott (2006) and Weld, Benartzi, Michaely, and Thaler (2009).  This is despite a dramatic deflation 

in the value of a dollar over the last century. In markets that are rising because of inflation or real 

growth, this average is maintained by splits. Weld et al. argue that standard explanations based on 

signaling or optimal trading ranges, which are most naturally thought of in real not nominal terms, 

are unable to explain the constancy of nominal prices, and several other related facts about active 

share price management. For example, both IPO prices and the share prices of open‐end mutual 

funds have also remained relatively constant. They propose instead that managers are simply 

following norms, adhering to an arbitrary historical convention from which there is no particular 

reason to deviate given investor expectations.  

Weld et al. study the stability of stock prices relative to the benchmark of no price 

management. Prices are not managed continuously, of course—on average and for individual 

stocks, prices are quite variable relative to the other extreme benchmark of a constant nominal 

price. Baker, Greenwood, and Wurgler (2009) study not the stationarity of average nominal prices 

but why they vary by a factor of two or more over time. 

Baker et al. propose that share prices are used as another tool to cater to time‐varying 

shareholder sentiment. In analogy to the dividend premium, they form a “low‐price premium” as 

the average market‐to‐book ratio of stocks whose prices fall in the bottom three deciles minus the 

Page 53: Behavioral Corporate Finance: An Updated Survey · Behavioral corporate finance, and behavioral finance more broadly, received a boost from the spectacular rise and fall of Internet

  49 

average of those with prices in the top three deciles. They find that when existing low‐price firms 

have high valuations, more firms split, and those splitters split to lower prices. IPOs also make for a 

powerful test, as they are free to list at almost any price. Consistent with catering, IPOs’ average 

prices vary closely with the low‐price premium. 

This leaves a question of interpretation. One derives from the strong cross‐sectional 

relationship between firm capitalization and nominal price. If shareholders take price as shorthand 

for size or growth potential, firms may split in order to “act small” when stocks that are small are 

especially highly valued. They cannot change capitalization, but they can change share price. 

2.6.5.   Executive compensation 

In the framework at the beginning of this section, we assumed that managers have the 

incentive to cater to short‐term mispricing. One question is why shareholders do not set up 

executive compensation contracts to force managers to take the long view.25 Bolton, Scheinkman, 

and Xiong (2005) suggest that short horizons may be an equilibrium outcome. They study the 

optimal incentive compensation contract for the dynamic speculative market of Scheinkman and 

Xiong (2003), in which two groups of overconfident investors trade shares back and forth as their 

relative optimism fluctuates. The share price in this market contains a speculative option 

component, reflecting the possibility that nonholders might suddenly become willing to buy at a 

high price. Bolton et al. find that the optimal contract may induce the CEO to take costly actions that 

exacerbate differences of opinion, thus increasing the value of the option component of stock prices, 

at the expense of long‐run value.  

                                                             

25 A separate but related question is how managers compensate lower‐level employees. Bergman and Jenter (2007) argue that rational managers may minimize costs by paying optimistic employees in overvalued equity, in the form of options grants. Benartzi (2001) offers a foundation for this sort of optimism, showing that employees have a tendency to extrapolate past returns, and as a consequence hold too much company stock. See also Core and Guay (2001) and Oyer and Schaefer (2005). 

Page 54: Behavioral Corporate Finance: An Updated Survey · Behavioral corporate finance, and behavioral finance more broadly, received a boost from the spectacular rise and fall of Internet

  50 

3.   Managerial Biases 

A second strand of behavioral corporate finance studies the behavior of irrational managers 

operating in efficient capital markets. By irrational managerial behavior we mean behavior that 

departs from rational expectations and expected utility maximization of the manager. We are not 

interested in rational moral hazard behavior, such as empire building, stealing, or plain slacking off. 

We are concerned with situations where the manager believes that he is actually close to 

maximizing firm value—and, in the process, some compensation scheme—but is in fact deviating 

from this ideal.26 We begin with a quick overview of the relevant psychology, then develop a simple 

theoretical framework, and follow with a review of this literature.  

3.1.  Background on managerial behavior 

The psychology and economics literatures relevant to understanding managerial behavior 

are vast. For us, the main themes are that individuals do not always form beliefs logically, nor do 

they convert a given set of beliefs into decisions in a consistent and rational manner. These recall 

the definitions of investor sentiment and irrational behavior that are assumed in market timing and 

catering studies. Following a note about corporate governance, we introduce and motivate the 

biases and nonstandard preferences that have been investigated in the context of managerial 

decisions. 

3.1.1.   Limited governance 

For less‐than‐fully‐rational managers to have an impact, corporate governance must be 

limited in its ability to constrain them into making rational decisions. This is analogous to the 

requirement of limited arbitrage for the market timing approach.  

                                                             

26 Our focus is on corporate finance. Camerer and Malmendier (2009) discuss the impact of less than fully rational behavior on other parts of organizations. 

Page 55: Behavioral Corporate Finance: An Updated Survey · Behavioral corporate finance, and behavioral finance more broadly, received a boost from the spectacular rise and fall of Internet

  51 

Assuming limited governance is no less reasonable than assuming limited arbitrage. Indeed, 

in the US, a significant element of managerial discretion is codified in the business judgment rule. 

Takeover battles and proxy fights are notoriously blunt tools. Boards may be more a part of the 

problem than the solution if they have their own biases or are pawns of management. For instance, 

Gompers, Ishii, and Metrick (2003) find that firms that elected policies to diminish shareholder 

rights experience lower stock returns.  And unlike in a traditional agency problem, which arises out 

of a conflict of interest between managers and outside investors, standard incentive contracts have 

little effect, because an irrational manager may well think that he is maximizing value.  

It is obvious from casual observation that top managers “matter,” in that they have the 

power to make decisions that affect investment and financing policy and firm value. There is also 

systematic evidence. Bertrand and Schoar (2003) find that individual managers have investment 

and financing styles and preferences, possibly inherent and possibly based on beliefs shaped by 

beliefs, that they bring from previous to new employers. For example, CEOs that use bigger 

mortgages for their own home purchases also use more leverage in their firms (Cronqvist, Makhija, 

and Yonker (2011)), although part of this effect can be attributed to endogenous firm‐manager 

matching. Kaplan, Klebanov, and Sorensen (2011) find that certain executive ability characteristics 

are correlated with firm performance. As one might expect, the expression of individual managerial 

decisions is stronger when the CEO is powerful or, similarly, when governance is weaker (Adams, 

Almeida, and Ferreira (2005) and Cronqvist et al.).  

3.1.2.  Bounded rationality 

Perhaps the simplest deviation from the benchmark of full rationality goes by the name of 

bounded rationality, introduced by Simon (1955). Bounded rationality assumes that some type of 

cognitive or information‐gathering cost prevents agents from making fully optimal decisions. 

Boundedly‐rational managers cope with complexity by using rules of thumb that ensure an 

Page 56: Behavioral Corporate Finance: An Updated Survey · Behavioral corporate finance, and behavioral finance more broadly, received a boost from the spectacular rise and fall of Internet

  52 

acceptable level of performance and, hopefully, avoid severe bias. Conlisk (1996) reviews the older 

bounded rationality literature; see Gabaix (2011) for a more recent modeling approach. Bounded 

rationality offers a reasonably compelling motivation for the financial rules of thumb that managers 

commonly use. We note some of these and consider the distortions that they create.  

3.1.3.  Optimism, overconfidence and hubris 

Most research in the managerial biases literature has focused on the illusions of optimism 

and overconfidence. Illustrating optimism, Weinstein (1980) finds that subjects believe themselves 

more likely than average to experience positive future life events (e.g. owning own home, living 

past 80) and less likely to experience negative events (being fired, getting cancer). Illustrating 

overconfidence in one’s own skills, and possibly optimism as well, Svenson (1981) finds that 82% 

of a sample of students placed themselves among the top 30% safest drivers. 

There are good reasons to focus on these particular biases in a managerial setting. First, 

they are strong and robust, having been documented in many samples, including samples of actual 

managers (Larwood and Whittaker (1977), March and Shapira (1987), and Ben‐David, Graham, and 

Harvey (2010)). Second, managerial decisions tend to be highly complex, a setting where 

overconfidence is most pronounced, and idiosyncratic, which reduces the potential for debiasing 

through learning (Gervais (2010)). Third, these biases are also often fairly easy to integrate into 

existing models. Optimism is usually  modeled as an overestimate of a mean ability or outcome and 

overconfidence as an underestimate of a variance. In this fashion we model the consequences of 

optimism, below, and also note situations in which an alternative assumption of overconfidence 

could lead to different conclusions. 

Finally, overconfidence also leads naturally to more risk‐taking. Even if there is no 

overconfidence on average in the population of potential managers, those that are overconfident 

are more likely to perform extremely well (and extremely badly), placing them disproportionately 

Page 57: Behavioral Corporate Finance: An Updated Survey · Behavioral corporate finance, and behavioral finance more broadly, received a boost from the spectacular rise and fall of Internet

  53 

in the ranks of upper (and former) management. And even if an individual manager is born without 

bias, an attribution bias—the tendency to take greater responsibility for success than failure (e.g., 

Langer and Roth (1975))—may lead successful managers to become overconfident, as modeled in 

Gervais and Odean (2001). 

3.1.4.  More on reference dependence 

Reference points and anchoring are equally compelling psychological foundations, when 

compared to overconfidence, and offer some empirical advantages in identifying behavioral effects 

in corporate finance. Section 2.1 describes the psychological underpinnings of reference points and 

anchoring. These hold special interest within a firm. A firm is collection of implicit and explicit 

contracts between managers and employees, the firm and its customers, creditors, underwriters, 

shareholders, and other stakeholders. It is natural to think of these as forming reference points in 

negotiations, and determining ex post the satisfaction of the various parties. For example, whether 

the management is satisfied with the performance of its underwriters depends on their 

performance relative to a reference price. Whether shareholders are satisfied with a merger offer 

depends on the price relative to recent transaction prices; we will see specific evidence of this later. 

Hart (2008) uses reference points more broadly as the underpinning for a theory of the 

firm. Using contracts as reference points to which parties feel entitled is a substitute for the 

assumptions of incomplete contracts and ex post bargaining over the surplus that drive the results 

in Grossman and Hart (1986) and Hart and Moore (1990). Because we do not observe this sort of 

bargaining within real firms, the reference point approach may outlive the existing architecture of 

the property rights theory of the firm. So far, however, much of the empirical evidence is focused on 

narrower applications of reference point preferences. 

Page 58: Behavioral Corporate Finance: An Updated Survey · Behavioral corporate finance, and behavioral finance more broadly, received a boost from the spectacular rise and fall of Internet

  54 

For the moment, we use overconfidence, instead of reference points, as an example of an 

organizing framework in the next section. The section on behavioral signaling at the end of the 

survey will develop a model using reference points. 

3.2.   Theoretical framework 

The derivation below is in the spirit of Heaton (2002) and Malmendier and Tate (2005), 

modified to match the notation in the market timing and catering model as much as possible. We 

assume the manager is optimistic about the value of the firm’s assets and investment opportunities. 

He balances two conflicting goals. The first is to maximize perceived fundamental value. To capture 

this, we augment fundamental value with an optimism parameter , 

KKf ,1 , 

where f is increasing and concave in new investment K. Note that here, the manager is optimistic 

about both the assets in place (f can include a constant term) and new opportunities. Once again, if 

traditional market imperfections cause the Modigliani and Miller (1958) theorem to fail, financing 

may enter f alongside investment. 

The manager’s second concern is to minimize the perceived cost of capital. We assume here 

that the manager acts on behalf of existing investors, because of his own stake in the firm and 

fiduciary duty. This leads to a similar setup to the market timing objective in Section 2.2, except that 

an optimistic manager never believes there is a good time to issue equity. In particular, since the 

capital market is efficient and values the firm at its true fundamental value of f‐K, the manager 

believes that the firm is undervalued by f, and thus in selling a fraction of the firm e he perceives 

that existing, long‐run shareholders will lose 

,Kfe . 

Page 59: Behavioral Corporate Finance: An Updated Survey · Behavioral corporate finance, and behavioral finance more broadly, received a boost from the spectacular rise and fall of Internet

  55 

Putting the two concerns together, the optimistic manager chooses new investment and 

financing to solve 

,,1max,

KfeKKfeK

We do not explicitly include a budget constraint. Instead, again to keep the notation simple, we 

consider its reduced‐form impact on f.  

Differentiating with respect to K and e gives the optimal investment and financial policy of 

an optimistic manager operating in efficient capital markets: 

eKf K

11

1, , and 

,,,1 KefKfKf ee . 

The first condition is about investment policy. Instead of setting the marginal value created 

from investment equal to the true cost of capital, normalized to be one here, managers overinvest, 

to the point where the marginal value creation is less than one. The more optimistic () is the 

manager and the less equity (e) he is forced to raise in financing investment, the greater the 

problem. To the extent that the manager has to raise capital by issuing equity, the cost of capital is 

scaled up by the same factor as the manager’s over‐optimism scales up the marginal product of 

capital, so raising equity offsets the distortion in investment caused by over‐optimism.  If 100% of 

the capital is raised by issuing equity, for example, investment is first‐best. The second condition is 

about financing. The marginal value lost from shifting the firm’s current capital structure away 

from equity is weighed against the perceived market timing losses. As in the analysis of irrational 

investors, we consider some special cases. 

Page 60: Behavioral Corporate Finance: An Updated Survey · Behavioral corporate finance, and behavioral finance more broadly, received a boost from the spectacular rise and fall of Internet

  56 

Investment policy. If there is no optimal capital structure, so that fe is equal to zero, the 

manager will not issue equity, setting e to zero, and there is no interaction among financing, 

internal funds, and investment. In this case, the optimistic manager will clearly overinvest: fK is less 

than unity. In Heaton (2002) and Malmendier and Tate (2005), there is an optimal capital structure, 

or more precisely an upper bound on debt. If the manager needs equity to invest (here, fe greater 

than zero), the degree of overinvestment falls. 

Needing equity is akin to having little cash or cash flow available for investment. Thus in 

this setup, investment can be strongly related to current cash flow and profits, controlling for 

investment opportunities. This leads to a behavioral foundation for the Jensen (1986) agency costs 

of free cash flow. But instead of receiving private benefits of control, managers are simply 

optimistic and overinvest from current resources as a result. Leverage reduces the degree of 

overinvestment by increasing fe, thereby increasing equity issues e and reducing K. 

In a more complex specification, these conclusions may change. One might have the 

manager optimistic only about assets in place, in which case there is no overinvestment, and there 

will typically be underinvestment as a firm approaches its debt capacity. Also, it is worth 

emphasizing that we are examining optimism in isolation here. Layering on other imperfections, 

such as risk aversion, may mean that optimism moves investment from an inefficiently low level 

toward the first best, as in Gervais, Heaton, and Odean (2010) and Goel and Thakor (2002). We will 

revisit some of these interactions when we discuss executive compensation. Hackbarth (2009) 

discusses another setting in which multiple biases can work in opposition, arguing that the 

combination of managerial optimism and overconfidence can reduce the underinvestment due to 

debt overhang (Myers (1977)). 

Financial policy. An optimistic manager never sells equity unless he has to. If there is an 

upper bound on leverage (fe greater than zero, here), optimism predicts a pecking order of 

Page 61: Behavioral Corporate Finance: An Updated Survey · Behavioral corporate finance, and behavioral finance more broadly, received a boost from the spectacular rise and fall of Internet

  57 

financing decisions: The manager relies on internal capital and debt and uses outside equity only as 

a last resort. Again, other imperfections may mitigate the aversion to equity. If the manager is risk 

averse with an undiversified position in the firm’s equity, for example, he may wish to issue equity 

even though it is below what he thinks it to be worth. 

Managerial overconfidence can have different effects on capital structure than optimism, 

Hackbarth (2009) argues. If overconfidence is modeled as underestimating the risk of earnings, 

managers may view their debt as undervalued and too expensive as a source of capital. The 

convexity of equity, on the other hand, leads managers to view their equity as overvalued. This 

reverses the pecking order that obtains under optimism. Suffice to say that theoretical predictions 

about the effect of optimism and overconfidence on capital structure are somewhat sensitive to the 

modeling framework.  

Other corporate decisions. It is not as easy to incorporate other decisions into this 

framework. Consider dividend policy. If the manager is more optimistic about future cash flow and 

assets in place than outside investors, he might view a dividend payment as more sustainable. On 

the other hand, if he views future investment opportunities, and hence funding requirements, as 

greater, he might be reluctant to initiate or increase dividends and retain internal funds instead. 

This analysis requires a more dynamic model of investment and cash flow and a decomposition of 

firm value into assets in place and growth opportunities.  

3.3.   Empirical challenges 

If the main obstacle to testing the irrational investors approach is finding a proxy for 

misvaluation, the challenge here is to identify optimism, overconfidence, or the behavioral bias of 

interest. Without an empirical measure, the irrational managers approach is typically difficult to 

distinguish from standard agency theory. That is, in Stein (2003), an empire‐building manager will 

Page 62: Behavioral Corporate Finance: An Updated Survey · Behavioral corporate finance, and behavioral finance more broadly, received a boost from the spectacular rise and fall of Internet

  58 

ecKKfeK

1max,

where  reflects the preference for or the private benefits that come with presiding over a larger 

firm, as in Jensen and Meckling (1976) or Grossman and Hart (1988), rather than optimism. 

Rational investors recognize the agency problem up front, so c reflects the cost of raising outside 

equity, and management and existing shareholders bear the agency costs. 

This reduced form is almost identical to the objective function of an optimistic manager. 

Both can generate overinvestment, underinvestment, cash flow‐investment sensitivities, pecking 

order financing, and so forth. Moreover, Stein points out that the agency model is itself hard to 

distinguish from models of costly external finance built on asymmetric information. Thus, to test 

the behavioral theories, one must separate the  related to overconfidence and optimism from the  

that arises from agency or asymmetric information problems. 

3.4.   Investment policy 

Despite the difficulty of obtaining direct, manager‐level measures of optimism and 

overconfidence, evidence is accumulating that these biases do affect business investment. 

3.4.1.   Real investment 

  The evidence does suggest that entrepreneurial startups are often made under a halo of 

overconfidence and optimism. Cooper, Woo, and Dunkelberg (1998) find that 68% of 

entrepreneurs think that their startup is more likely to succeed than comparable enterprises, while 

only 5% believe that their odds are worse, and a third of entrepreneurs view their success as all but 

guaranteed. The survey of French entrepreneurs by Landier and Thesmar (2009) gives the same 

message: At startup, 56% expect “development” in the near future while only 6% expect “difficulty.” 

Page 63: Behavioral Corporate Finance: An Updated Survey · Behavioral corporate finance, and behavioral finance more broadly, received a boost from the spectacular rise and fall of Internet

  59 

   The actual performance of startup investments is more sobering. Landier and Thesmar find 

that when surveyed three years into their endeavor, only 38% of French entrepreneurs expect 

further “development” while 17% anticipate “difficulty.” Leaving profitability aside entirely, only 

half of all startups survive more than three years (Scarpetta, Hemmings, Tressel, and Woo (2002)). 

Moskowitz and Vissing‐Jorgensen (2002) argue more generally that the return on private equity in 

the US between 1952 and 1999 is lower than seems justified given the undiversified nature of 

entrepreneurial investment. As a whole, the evidence on startup investments seems consistent with 

the overconfidence that Camerer and Lovallo’s (1999) experimental subjects display when making 

entry decisions.  

Optimism also may influence investment in more mature firms. Merrow, Phillips, and Myers 

(1981) compare forecast and actual construction costs for pioneer process plants in the energy 

industry. There is a strong optimism bias in project cost forecasts, with actual costs typically more 

than double the initial estimates. Statman and Tyebjee (1985) survey several other studies of this 

sort, involving military hardware, drugs, chemicals, and other development projects, and conclude 

that optimistic biases in cost and sales forecasts are fairly widespread.  

Malmendier and Tate (2005) perform cross‐sectional tests of the effects of optimism on 

investment. They form a manager‐level proxy for optimism based on the propensity for a manager 

to voluntarily hold in‐the‐money stock options in his own firm. Their intuition is that since the 

CEO’s human capital is already so exposed to firm‐specific risk, voluntarily holding in‐the‐money 

options is a strong vote of optimism.27 Using this optimism proxy for a large sample of US firms 

                                                             

27 Malmendier and Tate find that the propensity to voluntarily retain in‐the‐money options is not significantly related to future abnormal stock returns, supporting their assumption that such behavior indeed reflects optimism rather than genuine inside information. Sen and Tumarkin (2010) model the CEO’s portfolio choice and option exercise problem in more detail and argue that a more robust measure of optimism is simply whether the CEO sells or retains the shares received upon exercise. See Gider and Hackbarth (2010) for an overview of optimism and overconfidence proxies.  

Page 64: Behavioral Corporate Finance: An Updated Survey · Behavioral corporate finance, and behavioral finance more broadly, received a boost from the spectacular rise and fall of Internet

  60 

between 1980 and 1994, Malmendier and Tate find that the sensitivity of investment to cash flow is 

higher for the more optimistic CEOs. It is especially high for optimistic CEOs in equity‐dependent 

firms, that is, in situations where perceived financial constraints are most binding. Their results 

support the predictions of the basic optimism model.  

Ben‐David, Graham, and Harvey (2010) test whether survey‐based measures of 

overconfidence and optimism help to explain the level of investment as opposed to its sensitivity to 

cash flow. They ask financial executives to estimate the mean and variance of their firm’s stock 

return. This allows them to form separate optimism and overconfidence measures. A striking result 

is that financial executives are, indeed, extremely overconfident: their subjective 80% confidence 

intervals about the firm’s one‐year stock return contains the realized return only 33% of the time. 

They also connect these measures to the level of investment, and find that both optimism and 

overconfidence are associated with higher investment. 

  One category of investment that would seem particularly inviting to overconfident 

managers is research and development, where the payoff is inherently quite uncertain. Hirshleifer, 

Low, and Teoh (2010) find that overconfident managers—measured using options‐based proxies, 

as above, and the character of descriptions of the CEO in the press, similar to Malmendier and Tate 

(2004)—invest more in R&D and translate this to higher patent and patent citation count. At the 

same time, there is little relationship between their overconfidence measures and financial or stock 

market performance.  

In addition to the evidence above, keep in mind that optimism, as discussed earlier, shares 

many predictions with more established theories, and thus is a candidate to explain various earlier 

results. For example, the fact that managers invest rather than pay out cash windfalls (Blanchard, 

Lopez de Silanes, and Shleifer (1994)) looks like a moral hazard problem, but is also consistent with 

optimism. Likewise, some investment patterns that look like adverse‐selection‐driven costly 

Page 65: Behavioral Corporate Finance: An Updated Survey · Behavioral corporate finance, and behavioral finance more broadly, received a boost from the spectacular rise and fall of Internet

  61 

external finance may simply reflect a mistaken managerial belief that external finance is costlier. A 

possible example is the higher investment‐cash flow sensitivities of younger entrepreneurial firms 

(Schaller (1993)), which as noted above appear to be run by especial optimists. 

Moving away from optimism and overconfidence, a “bias” of bounded rationality appears to 

be a plausible explanation for some common capital budgeting criteria. For example, while the net 

present value criterion is the optimal capital budgeting rule (in efficient markets), real managers 

tend to employ simpler rules. Surveying practice in the 1970s, Gitman and Forrester (1977) find 

that less than 10% of 103 large firms use NPV as their primary technique, while over 50% use the 

IRR rule, which avoids a cost of capital calculation. The Graham and Harvey (2001) survey of CFOs 

also finds that the IRR rule is more widely used than NPV. Over 50% of CFOs actually use the 

payback period rule, an even less sophisticated rule that requires neither a cost of capital input nor 

cash flow forecasts beyond a cutoff date.  

Graham and Harvey also find that among managers who do use a discounting procedure 

tend to apply a firm‐wide discount rate rather than a project‐specific rate, again in contrast to 

normative principles.28  Kruger, Landier, and Thesmar (2011) suggest that this practice introduces 

significant investment distortions. Taking the project‐specific Capital Asset Pricing Model as a 

normative benchmark, Kruger et al. point out that multidivision firms that simply apply a weighted‐

average discount rate to all projects will overinvest in high beta divisions and underinvest in low 

beta divisions. Consistent with this prediction, they document that division‐level investment is 

positively related to the spread between the division’s market beta and the firm’s average beta. 

                                                             

28 A good question is whether the use of such rules is better understood as an agency problem than as bounded rationality. That is, executives might use simple rules to shorten the workday and save time for golf. However, Graham and Harvey find that high‐ownership managers are if anything less likely to use NPV and more likely to use the payback period rule. 

Page 66: Behavioral Corporate Finance: An Updated Survey · Behavioral corporate finance, and behavioral finance more broadly, received a boost from the spectacular rise and fall of Internet

  62 

  Loss aversion has also appeared as an explanation for certain investment patterns, such as 

in the widely asserted, but less documented, managerial propensity to “throw good money after 

bad.” Such behavior is most relevant for us to the extent that it reflects something more than 

rational career concerns, e.g. a situation where the manager tries to distort the updating process to 

maintain high compensation. Shefrin (2001) offers several anecdotes concerning major corporate 

investments that have the flavor of good money after bad. Statman and Sepe (1989) find that the 

market reaction to the termination of historically unprofitable investment projects is positive, 

suggesting that investors recognize that executives have a tendency to continue poor projects. 

Related evidence comes from the Guedj and Scharfstein (2008) study of drug development 

decisions. They find that single‐product early stage firms appear highly reluctant to abandon their 

only viable drug candidates, even when the results of clinical trials are less than promising. Some 

combination of agency, managerial optimism, and a gambling‐to‐get‐back‐to‐even attitude seems 

like a plausible explanation for these results.  

3.4.2.   Mergers and acquisitions 

In a seminal contribution to behavioral corporate finance, Roll (1986) outlines a hubris‐

based theory of acquisitions. He suggests that successful acquirers may be optimistic and 

overconfident in their own valuation of deal synergies, and fail to properly account for the winner’s 

curse. Roll interprets the evidence on merger announcement effects, surveyed by Jensen and 

Ruback (1983), Andrade, Mitchell, and Stafford (2001), and Moeller, Schlingemann and Stulz 

(2005), as well as the lack of evidence of fundamental value creation through mergers, as consistent 

with this theory.  

Malmendier and Tate (2004) develop this argument and use their options‐based proxy for 

CEO optimism to test it. They find patterns consistent with optimism and overconfidence. First, 

optimistic CEOs complete more mergers, especially diversifying mergers, typically suggested as 

Page 67: Behavioral Corporate Finance: An Updated Survey · Behavioral corporate finance, and behavioral finance more broadly, received a boost from the spectacular rise and fall of Internet

  63 

being of dubious value. Second, optimism has its biggest effect among the least equity dependent 

firms—when managers do not have to weigh the merger against an equity issue that they, as 

optimists, would perceive as undervalued. Third, investors are more skeptical about bid 

announcements when they are made by optimistic CEOs. Schneider and Spalt (2010) find similar 

results, including that offer prices are higher, but acquirer announcement returns are lower, when 

the target has (had) skewed returns. The announcement returns evidence is consistent with the 

theme of irrational managers operating in efficient markets.29  

Managerial biases research has taken a Freudian turn with Aktas, de Bodt, Bollaert, and 

Roll’s (2010) study of CEO narcissism. They measure narcissism, a trait related to but distinct from 

overconfidence, as the ratio of first person singular pronouns to total first person pronouns used in 

CEOs’ transcribed speeches. Thusly‐defined narcissist CEOs are more likely to be acquirers, and 

more likely to have initiated their transactions. This is interpreted as consistent with the high‐

stakes activity required to maintain the narcissistic ego. Targets run by narcissists, meanwhile, 

secure higher bid premia. Aktas et al. speculate that this arises because narcissistic CEOs demand 

extra compensation for the loss of ego associated with losing control. 

If managerial biases affect decisions because governance is limited, cross‐sectional variation 

in governance may be useful for identifying the effect. Yermack (1996) finds that firms with smaller 

boards of directors have higher firm value; Kolasinski and Li (2010) find that small boards 

dominated by independent directors reduce the impact of CEO overconfidence on acquisition 

frequency. They use negative future returns on CEO purchases as ex post evidence of ex ante 

overconfidence.  

                                                             

29 For more anecdotal evidence on the role of hubris in takeovers, see Hietala, Kaplan, and Robinson (2003) and Shefrin (2000, chapter 16).  

Page 68: Behavioral Corporate Finance: An Updated Survey · Behavioral corporate finance, and behavioral finance more broadly, received a boost from the spectacular rise and fall of Internet

  64 

To be useful in empirical work, these governance mechanisms need to be exogenous. 

Unfortunately, as Hermalin and Weisbach (2003) and Harris and Raviv (2008) point out, these are 

typically endogenous to firm performance. Nonetheless, the predictions here are typically 

concerning coefficients on interaction terms, so the endogeneity problem could be mitigated. 

Reference point thinking, in particular involving the offer price, also plays a role in merger 

activity. An offer must be made at a premium to the target’s current price, and the most salient and 

specific such prices are recent peaks, such as the target’s 52‐week high. There are a number of ways 

such salient but economically unremarkable prices could enter the psychology of merger 

negotiations. Valuing a company is a subjective task, and valuing a combination is doubly so. One 

could easily imagine that recent peak prices serve as anchors in such calculations on both the 

bidder and the target side. The target may use peak prices as a starting point for negotiations. Or, 

target shareholders may resist selling at a “loss” to a recent peak, akin to a disposition effect.30  

Baker, Pan, and Wurgler (2011) find that deal participants do indeed focus on recent price 

peaks. There is a spike in the distribution of offer prices at the target’s 52‐week high and other 

historical peaks. Bidding‐firm shareholders react negatively to the component of the offer price that 

is driven by the 52‐week high, which suggests that they rationally view this portion as 

overpayment. The probability that an offer goes through increases discontinuously when the offer 

exceeds the 52‐week high. This is an important result in that it represents unusually clean evidence 

of the real effects of behavioral corporate finance.  

Finally, Baker et al. find that reference point thinking may help to explain why mergers and 

stock market valuations are positively correlated: the offer premium required to exceed a recent 

price peak is smaller when valuations have increased. Conversely, when valuations have                                                              

30 Note that while some of these effects involve managerial biases, others represent investor biases and thus the evidence below could also be included in our earlier sections about investor irrationality. 

Page 69: Behavioral Corporate Finance: An Updated Survey · Behavioral corporate finance, and behavioral finance more broadly, received a boost from the spectacular rise and fall of Internet

  65 

plummeted, targets may fail to adjust from prior peak anchors and, as a result, ask for valuations 

that are simply implausible to bidders.  

3.5.   Financial policy 

There is a growing body of evidence that managerial biases affect financing patterns. 

Existing work addresses the timing and pricing of equity issues, features of IPOs, capital structure 

and dividend policy, and financial contracting. Reference dependence plays a prominent role in 

these studies.  

3.5.1.  Equity issues 

Does the CEO drive a firm’s stock returns? If so, then a CEO would rightly be proud, and 

shareholders should take notice, when she has created value and raised the share price above the 

level that prevailed when she took the helm. If not, for example if share prices are dominated by 

aggregate moves, then that historical price does not serve as a particularly meaningful reference 

point for CEO‐specific value creation.  

Baker and Xuan (2011) find evidence that CEO‐specific share price performance does 

indeed affect financing activity. Equity issuance is responsive to recent stock returns, but 

considerably more so when they occur during the current CEO’s tenure. In particular, the 

probability of equity issuance in a follow‐on offering increases discontinuously when the share 

price exceeds the inherited price.  

Apparently, some market participants involved in equity issuance attribute “value creation” 

to the CEO and her team. To be clear, this by itself is not necessarily a behavioral phenomenon; the 

intriguing result is the effect of the inherited share price level even though subsequent market‐level 

movements beyond the CEO’s control complicate the attribution of value creation. The attribution 

error could be on the investor side, with management having to wait until this point in order to 

Page 70: Behavioral Corporate Finance: An Updated Survey · Behavioral corporate finance, and behavioral finance more broadly, received a boost from the spectacular rise and fall of Internet

  66 

convince investors that issue terms were appealing. Instead of that effect, or in addition to it, the 

management team may view crossing the inherited price threshold as an opportunity to time the 

equity market. 

3.5.2.  IPO prices 

IPO underpricing can also be understood from the perspective of reference‐point 

managerial preferences. Loughran and Ritter (2002) develop an explanation that combines 

reference‐point preferences and mental accounting (Thaler (1980, 1985)). An important facet of 

IPO pricing is that the investment bankers underwriting the offering form an initial file price range, 

as they shop the deal to institutional investors. If demand for the new shares is high, the bankers 

will price the offering at the high end of this range. If it is low, they will price the offering at the 

midpoint, or sometimes lower. On the first day, prices are a market outcome of the new supply and 

demand.  

Loughran and Ritter assume that issuing managers mentally account for two quantities in 

judging an offering’s success: the (perceived) gain from the gap between the first day closing price 

and a natural reference point, the midpoint of the file price range; and the (real) loss from the 

dilutive effect of the underpricing. If the gain is judged to outweigh the loss, where each is evaluated 

separately with the prospect theory value function, the executives are satisfied. Intuitively, they 

may be too overwhelmed by the “windfall gain” to complain much about underpricing.31 

This setup is designed, in part, to explain the pattern that underpricing is greater when the 

offer price is above the initial file price range. Loughran and Ritter (2002) find that in issues where 

the offer price is below the minimum of the file price range, first‐day returns are a relatively small 

                                                             

31 Loughran and Ritter assume that the underwriter prefers underpricing, perhaps because it generates profitable rent‐seeking activities among investors, e.g. trading with the underwriter’s brokerage arm, or because it reduces marketing costs. 

Page 71: Behavioral Corporate Finance: An Updated Survey · Behavioral corporate finance, and behavioral finance more broadly, received a boost from the spectacular rise and fall of Internet

  67 

4%, on average, while those priced above the maximum have average first‐day returns of 32%. This 

is consistent with issuers acquiescing in severe underpricing only when they are simultaneously 

getting good news in the form of upward revisions from the filing range.32  

Ljungqvist and Wilhelm (2005) test some of the behavioral underpinnings of the Loughran 

and Ritter view. Using data on the ownership stakes of executives in IPO firms, they crudely proxy 

for the proposed notion of issuer satisfaction by taking the dollar amount of executives’ perceived 

“gain” from revisions from the midpoint of the file price range and subtracting the dollar amount of 

dilution due to underpricing. They find that executive teams that are more “satisfied” with their 

IPOs by this criterion are more likely to use the same underwriter for seasoned offerings, and to 

pay higher fees for those transactions. 

3.5.3.   Raising debt 

  Borrowers and lenders use past terms as anchors or reference points for current terms. 

Dougal, Engelberg, Parsons, and Van Wesep (2011) find that the nominal level of historical 

borrowing costs exerts a strong influence on the time t cost of debt, controlling for a variety of time 

t borrower characteristics. The effect appears for all credit rating categories. For example, firms 

that took out credit from a banking syndicate between 2005 and 2007 saw the influence of the 

2008 financial crisis have a muted impact on their 2008 borrowing costs from the same source. For 

firms whose credit rating remained constant over this period, one‐third received exactly the same 

borrowing rates as in the pre‐crisis period. Comparable firms that hadn’t established such anchor 

terms saw higher borrowing costs. 

                                                             

32 See Benveniste and Spindt (1989) for an alternative explanation for this asymmetry based on information gathering in the book‐building process, and Edelen and Kadlec (2003) for one based on sample truncation bias related to the withdrawl of IPOs whose prospects deteriorate during the waiting period. 

Page 72: Behavioral Corporate Finance: An Updated Survey · Behavioral corporate finance, and behavioral finance more broadly, received a boost from the spectacular rise and fall of Internet

  68 

  It is easy to understand how prior terms are natural starting points for thinking about and 

negotiating new terms. The need for a fixed starting point could be particularly high in periods of 

dramatic change in the financial environment. Dougal et al. find additional patterns that further tie 

their results to anchoring: specific managers and bankers appear to form relationships that are 

most affected by the bias; when a firm changes lead banks, the effect of past terms deteriorates; 

and, when a firm changes CEO or CFO, the effect of past terms deteriorates. 33  

  Similar to how reference point prices affect merger activity or manager‐specific reference 

point prices on equity issues, this experiment provides further evidence that highly sophisticated 

actors—in this case, managers, bankers, and investors jointly—are unable to “integrate out” the 

past. Future research may better identify the real effects of this. A natural hypothesis, for example, 

is that borrowers who are being offered a deal because of the happenstance of favorable past terms 

will raise more and invest more. 

3.5.4.   Capital structure 

The most basic optimism model predicts a pecking order financing policy, as pointed out by 

Heaton (2002). Thus, much of the existing evidence of pecking‐order policies, from Donaldson 

(1961) to Fama and French (2002), is at face value equally consistent with pervasive managerial 

optimism. And the notion of pervasive managerial optimism does not seem farfetched. In Graham’s 

(1999) survey, almost two‐thirds of CFOs state their stock is undervalued while only three percent 

state it is overvalued. Such responses are all the more striking given that the survey was taken 

shortly before the Internet crash. 

To distinguish optimism from other explanations of pecking order behavior, such as 

adverse selection as in Myers and Majluf (1984), a natural test would use cross‐sectional variation 

                                                             

33 The authors argue that costly renegotiation of terms cannot explain these results.  

Page 73: Behavioral Corporate Finance: An Updated Survey · Behavioral corporate finance, and behavioral finance more broadly, received a boost from the spectacular rise and fall of Internet

  69 

in measured optimism to see whether such behavior is more prevalent in firms run by optimists. To 

our knowledge, exactly this test has not been conducted, but certain results in Malmendier and Tate 

(2004, 2005) have a related flavor. First, and as noted above, firms run by optimists (as identified 

by their options‐based proxies for optimism) display a higher sensitivity of investment to internal 

cash flow. Second, managers classified as optimistic show a differentially higher propensity to make 

acquisitions when they are not dependent on external equity.  

  Bounded rationality also makes an appearance in financial policy in the form of the use of 

simple targets for capital structures and payouts. Graham and Harvey (2001) find that 10% of the 

CFOs in their sample use a “very strict” target debt‐equity ratio and 34% use a “somewhat tight” 

target or range. Such leverage targets are typically defined in terms of book values of equity and 

debt, and Welch (2004) confirms that market leverage is largely allowed to float with stock prices. 

Whether this is a rule of thumb, a boundedly rational focus on slower moving book values, or a 

rational recognition that book values are a better proxy for liquidation value than market value is 

hard to prove. Likewise, and as mentioned before, Lintner’s (1956) field interviews reveal a set of 

common rules of thumb in payout policy that lead to a reasonably accurate empirical specification 

for dividends. Brav, Graham, Harvey, and Michaely (2005) find that some of these rules still apply 

fifty years later.  

3.5.5.   Contracting and executive compensation 

  Landier and Thesmar (2009) examine financial contracting between rational investors and 

optimistic entrepreneurs.34 They highlight two aspects of contracting with optimists. First, because 

optimists tend to inefficiently persist in their initial business plan, the optimal contract transfers 

control when changes are necessary. (Kaplan and Stromberg (2003) find that contingent transfers 

                                                             

34 Manove and Padilla (1999) also consider how banks separate optimists and realists. They focus on the overall efficiency of the credit market.  

Page 74: Behavioral Corporate Finance: An Updated Survey · Behavioral corporate finance, and behavioral finance more broadly, received a boost from the spectacular rise and fall of Internet

  70 

of control are common features of venture capital contracts.) Second, because optimists believe 

good states to be more likely, they are willing to trade some control and ownership rights in bad 

states for greater claims in good ones; in this sense, the optimal contract “pays the entrepreneur 

with dreams.” Ultimately, optimists may self‐select into short‐term debt, as it transfers payments 

and control to the investor in states that they think are unlikely, while realistic entrepreneurs 

prefer less risky long‐term debt.  

Landier and Thesmar find some empirical evidence of this separation in data on French 

entrepreneurs. Among other results, they find that the use of short‐term debt is positively related 

to an ex post measure of optimistic expectations, the difference between realized growth and initial 

growth expectations. They also find that the use of short‐term debt is positively related to 

psychologically‐motivated instruments for expectations, such as regional sunlight exposure and 

rates of mental depression. 

  Some related phenomena appear in the context of biased executives’ compensation 

contracts. Standard contracting models seem unable to explain basic aspects of CEO compensation. 

For example, Hall and Murphy (2002) and Dittman and Maug (2007) point out that convex 

incentives are commonly induced through stock options. Yet these turn out to be hard to calibrate 

to standard models with risk‐neutral shareholders and risk‐averse, undiversified executives. 

Dittman and Maug argue that such setups actually tend to predict negative base salaries.  

  Gervais, Heaton, and Odean (2010) derive the optimal compensation contract for a risk‐

averse but overconfident manager. The manager overweights his private information, so the 

optimal contract balances the standard issue of overcoming his risk aversion with the need to avoid 

rash investments.  The most basic effect is that if the manager is highly overconfident, shareholders’ 

wealth‐maximizing contract is highly convex, because the manager overvalues it. This effect is 

reminiscent of paying with dreams. 

Page 75: Behavioral Corporate Finance: An Updated Survey · Behavioral corporate finance, and behavioral finance more broadly, received a boost from the spectacular rise and fall of Internet

  71 

  The prospect theory value function provides another explanation for stock options and 

positive base salaries as optimal contracts. Dittman, Maug, and Spalt (2010) show that implausible 

parameters are not required; for example, the manager’s reference wage can be close to last year’s 

salary and bonus. The manager’s risk tolerance is near zero around the reference point but 

increases rapidly as payout increases. This necessitates high‐powered, convex contracts even with 

optimal risk sharing.  This is consistent with high salaries and positive stock and stock option 

holdings that we observe.  

4.   Behavioral Signaling 

  Another behavioral approach to corporate finance is in an embryonic stage. We include it 

alongside more mature research frameworks because of its theoretical distinctiveness and seeming 

promise. We also happen to find it interesting ourselves; our discussion here will focus on Baker 

and Wurgler (2011). The model involves quasi‐rational investors, so in a conceptual sense it falls 

between the market timing and catering research, which assumes irrational investors, and the 

managerial biases research, which assumes fully rational investors.   

  The core idea of signaling models since Spence (1973) is that “good” types can separate 

themselves by taking some action that is less costly for them than it is for “bad” types. In corporate 

finance, classic applications include the capital structure models of Leland and Pyle (1977), Ross 

(1977), and Myers and Majluf (1984); the dividend models of Bhattacharya (1979), John and 

Williams (1985), and Miller and Rock (1985); the convertible debt model of Harris and Raviv 

(1985); and, the IPO underpricing models of Allen and Faulhaber (1989), Grinblatt and Hwang 

(1989), and Welch (1989). Although the nature of the signaling mechanism varies, all of these 

models feature participants with standard preferences and rational expectations. 

Page 76: Behavioral Corporate Finance: An Updated Survey · Behavioral corporate finance, and behavioral finance more broadly, received a boost from the spectacular rise and fall of Internet

  72 

  The defining characteristic of behavioral signaling models is that the signaling mechanism 

derives from nonstandard preferences or judgmental biases. The model of dividend policy we 

discuss below is an example. It relies on prospect theory preferences and narrow framing. 

4.1.  Theoretical Framework 

  There is no standard theoretical framework to outline here at this time. Indeed, there are 

many behavioral distortions one could imagine basing a signaling model upon, and each might have 

a somewhat different implementation and application. We will review a specific model of dividend 

signaling based on Baker and Wurgler (2011).  

The main goal of this dividend‐signaling model is, as usual, to explain why firms pay 

dividends at all. Secondary goals are to shed light on other facts of dividend policy. These include 

the fact that dividends are often not raised or lowered for long stretches; that dividend cuts are 

greeted very negatively; and, that dividends can be described using the Lintner (1956) partial‐

adjustment model. We outline the model and then return to more detailed empirical implications.  

The signaling mechanism is based on nonstandard investor preferences, not willful 

destruction of firm value through investment distortions or taxes. In particular, it is based on the 

reference dependence and loss aversion features of the prospect theory value function of 

Kahneman and Tversky (1979). Reference dependence refers to the propensity to judge utility 

based on losses and gains relative to a context‐specific reference point. Loss aversion denotes the 

tendency to perceive more disutility from losses than utility from equal‐size gains. Suffice it to say 

that a great deal of research from psychology and economics supports these effects—see, e.g., 

Kahneman (2003). 

The model’s first key ingredient is that a reference point level of dividends appears in the 

investor’s objective function. Per loss aversion, there is a kink in utility, so that the negative effect of 

Page 77: Behavioral Corporate Finance: An Updated Survey · Behavioral corporate finance, and behavioral finance more broadly, received a boost from the spectacular rise and fall of Internet

  73 

a $0.01 drop in dividends just below the reference point is greater than the positive effect of a $0.01 

increase in dividends just above. The second key ingredient is that the manager cares about the 

current estimate of firm value as well as the long‐term welfare of investors. 

The model focuses on two periods: t = 1 and 2. There are two players: a benevolent 

manager and an investor to whom dividend cuts from the current reference point level are 

discontinuously painful. In the first period, the investor arrives with an exogenous reference point 

d*. The manager also receives private information about cash earnings 1 and pays a dividend d1 in 

the first period. Given this dividend, the investor learns something about the manager’s private 

information and hence the value of the firm. This dividend, which may be below, above, or equal to 

d*, in turn forms a new reference point for the liquidating dividend d2. In some ways, this model can 

be viewed as a snapshot of a multi‐period model.  

In this model, reference points shape dividend policy in multiple ways. On the one hand, to 

the extent that today’s dividend is the reference point against which future dividend payments will 

be judged, the manager would like to restrain current dividends, saving some resources for the next 

period to make up for a possible shortfall in future earnings. On the other hand, setting aside effects 

on future investor welfare, the manager would like to pay a dividend today that exceeds the current 

reference point. Moreover, because the manager also cares about the current estimate of firm value, 

which for simplicity we take to mean the estimate of first period cash earnings, he might also 

increase dividends beyond the current reference point to signal private information about the 

firm’s ability to pay. This signaling mechanism works because firms with limited resources are 

unwilling to incur the expected future cost of missing an endogenous reference point. Coming back 

to the formalities, we have: 

Page 78: Behavioral Corporate Finance: An Updated Survey · Behavioral corporate finance, and behavioral finance more broadly, received a boost from the spectacular rise and fall of Internet

  74 

Manager utility. The manager cares about what the investor thinks about 1 today because 

that determines today’s stock price. He also cares about the investor’s long run utility. The 

simplified objective function is: 

*211 |,1 ddduEE im ,   

where d1 and d2 are the period‐specific dividends of the firm, u is the investor’s utility function, 

given an exogenous initial reference point of d*, and Em and Ei are the expectations operators for the 

manager and the investor, respectively.35 

Investor utility. The manager’s objective is standard. The interesting aspect of this 

signaling model is that the investor has a kink in his preferences for dividends d1 and d2. The first 

kink is around an exogenous reference point for first period dividends d* and the second kink is 

around an endogenous reference point for second period dividends: 

u d1, d2 | d* d1 b d1 d * d1 d * d2 b d2 d1 d2 d1 . 

In other words, the investor cares about fundamental value, or total dividend payments, but 

with a twist. The level of the reference point comes from historical firm dividend policy, and b is 

greater than zero to incorporate loss aversion. This utility function is in the spirit of prospect theory 

with a kink at a reference point. The second period reference point equals first period dividends d1 

by assumption. In reality, the reference point and the intensity of the reference point b may be 

determined by a long history of levels and changes in dividend policy. The fact that each dividend 

payment forms a separate reference point also requires narrow framing. This is not a reference 

                                                             

35 The fact that the investor’s expectation of 1 appears directly into the manager’s objective is an innocuous assumption, because in equilibrium the stock price will be a linear transformation of this expectation. 

Page 79: Behavioral Corporate Finance: An Updated Survey · Behavioral corporate finance, and behavioral finance more broadly, received a boost from the spectacular rise and fall of Internet

  75 

point applied to total ending wealth, but much more narrowly both across stocks and time, in the 

spirit of Barberis, Huang, and Thaler (2006). 

Information. For simplicity, the manager has no control over the cash earnings of the firm. 

This is a bit different from a traditional signaling model where the manager must destroy firm value 

to impress the capital markets. There is also no agency problem; the manager is not able to keep the 

cash for himself, and no real value is created or destroyed with dividend policy. The fundamental 

value of the firm appears in two installments and totals  21 . Think of these as cash earnings that 

are not observable to the investor. This is an extreme assumption of asymmetric information that 

highlights the intuition. For simplicity, assume that the second‐period cash earnings have a uniform 

distribution,  2,0~2 U .  

Budget constraint. There is no new equity or debt available to finance the payment of 

dividends and no excess cash balances available in the first period. The most the manager can pay 

in the first period is 1, and the most he can pay in the second period is 2 plus any savings from the 

first period. Given a benevolent manager and the absence of new financing, this implies constraints: 

110 d  and  1212 dd .   

Equilibrium. Combining the above, there are three important effects that appear in the 

manager’s objective function. First, there is sometimes an advantage to paying out dividends 

immediately. Consider a first period dividend below the reference point d*. Setting aside the effect 

on the second period reference point, these dividends will be valued on the margin at b+1 times the 

payout, instead of simply the payout. Above d*, there is no marginal benefit from merely shifting 

payout from the second period forward. Second, by increasing the dividend today, the investor’s 

estimate Ei[1] of the unobservable cash earnings rises through an equilibrium set of beliefs that 

map dividend policy to cash earnings. This enters into the manager’s utility function directly. Third, 

Page 80: Behavioral Corporate Finance: An Updated Survey · Behavioral corporate finance, and behavioral finance more broadly, received a boost from the spectacular rise and fall of Internet

  76 

increasing the dividend in the first period, for either of these rationales, produces an expected 

future cost to investor utility that comes from the possibility of falling short of the reference point 

set for the second period.  

These three motivations combine to simplify the manager’s utility function: 

1 b d1 d * d1 d * Ei 1 | d1 1 b d1 1

2

2

d1 1

2

The first term reflects striving to avoid falling short of the initial reference point. The second term 

reflects concern about share price. The third term reflects the expected cost of falling short of a new 

reference point; there is no cost if the manager adopts a very conservative dividend policy of paying 

half of first‐period earnings. Given the uniform distribution of 2, the expected cost is quadratic as 

dividends rise from this point and increasing in the intensity of the reference point.  

Intuitively, these considerations suggest three ranges of dividend policies in equilibrium. 

There is a high payout ratio for firms with the extra motivation due to signaling to clear the initial 

reference point of d*. Next, managers cluster at d* once this marginal effect drops out, i.e. they 

maintain their existing dividend level exactly. Finally, there is a lower payout ratio for firms with 

first‐period earnings well above the initial reference point. These lucky firms nonetheless pay 

higher dividends to separate themselves from each other and from the pool at d*. Specifically, there 

exists an equilibrium where: 

d1 = 1 if 1 < d*,  

= d* if d* < 1 < *, and 

=  12 1

1 1b  if 1 > *, 

Page 81: Behavioral Corporate Finance: An Updated Survey · Behavioral corporate finance, and behavioral finance more broadly, received a boost from the spectacular rise and fall of Internet

  77 

with * satisfying  12 *d * 1 b

1 1b 2 d * 1

2 * 2 0 ,  

and equilibrium beliefs of:  

Ei[1 | d1] = d1 if d1 < d*, 

= **21 d if d1 = d*, and 

= 2 d1 1 1

b if d1 > 12 *

1 1b .  

There are intuitive comparative statics with respect to b, the cost of falling below the 

reference point. In this equilibrium, it can be shown that as b increases and  decreases, there is 

more clustering of dividends at the reference point d* (* increases), and the market reaction to d1 < 

d* increases, because there is more information revealed in a near miss.  

4.2.   Applications 

  We will discuss the empirical relevance of this dividend‐signaling model and then speculate 

a bit about potential future applications of behavioral signaling.  

4.2.1  Dividends 

An important feature of the reference points model is that it is consistent with what 

managers say about dividend policy.  In the Brav et al. (2004) survey, executives disavow the notion 

that they pay dividends because it destroys firm value and therefore signals strength. This is the 

basis, however, of numerous non‐behavioral signaling models. At the same time, managers do agree 

with the notion that dividends are a “signal” of some sort. The behavioral signaling model with 

dividends as reference points can signal financial soundness without burning money.  

Page 82: Behavioral Corporate Finance: An Updated Survey · Behavioral corporate finance, and behavioral finance more broadly, received a boost from the spectacular rise and fall of Internet

  78 

Behavioral signaling can also give foundations to the Lintner (1956) model, which has 

proved a difficult task using traditional approaches. In the equilibrium described above, firms with 

good earnings realizations (1 > *) follow a partial‐adjustment policy and are more generally 

smoothed relative to earnings. The Lintner model takes the previous dividend as the starting point 

for any adjustment in this period; the behavioral signaling model predicts that the dividend level 

will be constant for many firms and adjusted only when earnings are sufficiently extreme. On 

average for all firms, dividends increase less than one‐for‐one with earnings, consistent with partial 

adjustment.  

The reaction to dividend changes is asymmetric, with cuts being particularly painful 

(Aharony and Swary (1980)). Most standard signaling models do not incorporate this asymmetry. 

On the other hand, it immediately follows from a model with loss‐averse investors who use lagged 

dividends as a reference point—the relevant effect is that cutting dividends, even slightly, fully 

reveals that the firm’s financial strength is low.   

A fundamental theme of the model is that the level of dividends needs to be salient and 

memorable in order to maximize the strength of the signal. If investors don’t notice their dividend 

or don’t notice changes, the reference‐point mechanism fails. In fact, similar to what Thomas (1989) 

finds in earnings levels, dividend levels and changes tend also to be in easy‐to‐digest round 

numbers, such as multiples of five and ten cents per share. This feature of dividend policy again has 

no natural interpretation within traditional signaling theories.  

4.2.2.   Other applications 

  Earnings management presents another potential application for behavioral signaling. 

Important features of the reported earnings process are reminiscent of the dividend process. 

Burgstahler and Dichev (1997) and Degeorge et al. (1999) find that earnings are managed to meet 

or exceed salient reference points. As discussed earlier, these include zero earnings, lagged 

Page 83: Behavioral Corporate Finance: An Updated Survey · Behavioral corporate finance, and behavioral finance more broadly, received a boost from the spectacular rise and fall of Internet

  79 

earnings, and analyst expectations. In addition, reported earnings are smoothed versions of true 

earnings, involving a partial‐adjustment process not unlike the Lintner model.  

  A loss‐aversion based mechanism isn’t as natural in the earnings context, however. 

Reported earnings are less tangible and visible to the mass of investors than dividends; loss 

aversion to reported earnings per se is unnatural.36 A more realistic signaling mechanism might be 

based on irrational beliefs.  

For example, suppose that investors overreact if reported earnings fall below the threshold 

of prior earnings. (Skinner and Sloan (2002) find that growth firms, for which information opacity 

is highest and signaling most useful, do exhibit an asymmetric response to earnings surprises.) 

Reported earnings can then become a signal: Managers with favorable private information can 

aggressively manipulate earnings upward and establish higher reference points for future earnings. 

Distinguishing between two types of investors—noise traders with incorrect beliefs and 

arbitrageurs with rational expectations but limited capital and risk‐bearing ability—allows one to 

preserve a rational expectations equilibrium concept. In this setup, managers are essentially 

signaling to the arbitrageurs; the noise traders provide the mechanism.  

  Stock splits have also been modeled as signals in rational expectations frameworks, without 

clear success. The costly signaling mechanisms in Brennan and Copeland (1988) and Brennan and 

Hughes (1991) involve transaction costs: roughly speaking, firms split to lower‐priced shares to 

increase trading costs on their investors. Unfortunately, Baker and Powell (1993) survey managers 

and they say that splits are if anything an effort to improve liquidity. 

                                                             

36 On the other hand, Degeorge et al. propose that executives themselves may derive personal utility from meeting thresholds. 

Page 84: Behavioral Corporate Finance: An Updated Survey · Behavioral corporate finance, and behavioral finance more broadly, received a boost from the spectacular rise and fall of Internet

  80 

It is not hard to sketch a simple behavioral signaling model of splits that is more intuitive. 

For example, suppose that noise traders coarsely categorize low‐nominal‐price firms, all else equal, 

as growth firms (Baker, Greenwood, and Wurgler (2009)). In this environment, splitters can 

credibly separate themselves in the eyes of rational arbitrageurs because they know they can 

deliver higher earnings next period and not risk the wrath of the noise traders. Skinner and Sloan’s 

(2002) results are also compatible with this simple model.   

5.   Some open questions 

  The behavioral corporate finance literature has matured to the point where one can now 

sketch out a handful of canonical theoretical frameworks and use them to organize many dozens of 

empirical studies. Our review of this evidence indicates that behavioral approaches offer a useful 

complement to the other corporate finance paradigms. They deliver intuitive and sometimes quite 

compelling explanations for important financing and investing patterns, including some that are 

difficult to reconcile with existing theory.  

In its current state of flux, the field offers a number of exciting and important research 

questions. We close by highlighting just a few. In no particular order, we wonder: 

Are behavioral factors at the root of why managers do not more aggressively pursue the tax 

benefits of debt, as in Graham (2000)? Hackbarth (2009) develops a theoretical argument 

along these lines. 

While the existing literature has generally considered the two approaches separately, the 

irrational manager and irrational investor stories can certainly coexist. Would a model 

featuring a correlation between investor and managerial sentiment lead to new insights? 

Page 85: Behavioral Corporate Finance: An Updated Survey · Behavioral corporate finance, and behavioral finance more broadly, received a boost from the spectacular rise and fall of Internet

  81 

What other phenomena can be modeled with behavioral signaling? How can such models be 

tested? 

What are the determinants of managerial “horizons,” and how can they be measured and 

appropriately governed?  

To what extent should investment bankers be viewed as institutions whose business model 

is to identify and cater to emerging pockets of investor sentiment? 

To what extent should private equity funds be viewed as firms whose business model is to 

capitalize on equity and debt markets that are not fully integrated, with separate investor 

demand shocks and inconsistent pricing? 

What are the behavioral explanations for the recent financial crisis? Barberis (2011) starts 

to connect the dots. 

How is the banking system affected by inefficiencies in the capital markets? Should 

regulation aim to insulate banks from bubbles? Should this operate through broad capital 

regulations, or more narrowly? 

Are derivative instruments – most noteworthy in recent history, credit default swaps (CDS) 

and CDOs – prone to misvaluation? To what extent do they make corporate outcomes more 

efficient by lowering the ex ante cost of capital through efficient risk sharing or by 

predicting default? To what extent are they the source of mispricings that propagate into 

debt and equity prices? 

What determines investor sentiment, and how is it managed through corporate investor 

relations (Brennan and Tamarowski (2000))? Potential avenues to consider are interactions 

Page 86: Behavioral Corporate Finance: An Updated Survey · Behavioral corporate finance, and behavioral finance more broadly, received a boost from the spectacular rise and fall of Internet

  82 

with past stock market returns, technological change and the valuation of new industries, 

media coverage, financial analysts and financial reporting, and investment banking.  

Do equity and debt market timing reduce the overall cost of capital by a small amount or a 

large amount? Dichev (2004) offers an approach here. 

To what extent can features of financial contracts and securities be understood as a 

response to assorted behavioral biases? Williamson took first steps here. In the context of 

consumer contracts, Della Vigna and Malmendier (2004) suggest that credit cards and 

health club contracts are shaped by naïve expectations and time‐inconsistent preferences.  

What is the impact of investor inertia and limited attention on corporate finance? Baker, 

Coval and Stein (2007) and Della Vigna and Pollet (2009) consider stock swaps and the 

timing of corporate disclosure. Hirshleifer and Welch (2002) develop implications for 

organizations.  

How should one approach the proper regulation of inefficient markets and financial 

reporting? The financial crisis has generated discussion about the role of the Fed and the 

SEC with regard to identifying and managing investor sentiment and bubbles.  

What are the limits of corporate arbitrage, including detecting and generating mispricing, 

maintaining reputation, and avoiding fraud? 

Can a catering approach help to explain the diversification and subsequent re‐focus wave 

that has taken place in the US since the late‐1960s?  

How significant is the economy‐wide misallocation of capital caused by collected behavioral 

distortions, and in particular how do these distortions interact with traditional capital 

market imperfections? For example, if there is underinvestment due to agency or 

Page 87: Behavioral Corporate Finance: An Updated Survey · Behavioral corporate finance, and behavioral finance more broadly, received a boost from the spectacular rise and fall of Internet

  83 

asymmetric information, bubbles may bring investment closer to the efficient level—or 

overshoot.  

If bounded rationality or investor pressures lead managers to rely on specific performance 

metrics, will third parties exploit this? The marketing of takeovers and financing vehicles as 

EPS‐improving transactions by investment banks is a potential example. More generally, 

what profit opportunities are created by behavioral biases of investors and managers?  

To what extent are corporate “hedging” policies actually directional bets? The evidence in 

Brown, Crabb, and Haushalter (2002) and Faulkender (2005) suggests that in many 

companies, interest rate risk management and the use of derivatives has little to do with 

textbook hedging.  

What are the normative, legal, and ethical implications of market‐driven corporate finance? 

Should managers be encouraged to respond to movements in prices and interest rates that 

do not reflect underlying fundamentals? Jensen (2005) explores the agency problems that 

arise from overvalued equity. 

In the Introduction, we pointed out that the normative implication of assuming irrational 

investors is to insulate managers from short‐term market pressures, while the implication 

of assuming irrational managers approach is to obligate them to follow market prices. What, 

in the real world, is the right balance between discretion and market pressure? 

Page 88: Behavioral Corporate Finance: An Updated Survey · Behavioral corporate finance, and behavioral finance more broadly, received a boost from the spectacular rise and fall of Internet

  84 

References 

Acharya, Viral, and Lasse Pedersen, 2005, Asset pricing with liquidity risk, Journal of Financial Economics 77, 375‐410. 

Adams, Renee, Heitor Almeida, and Daniel Ferreira, 2005, Powerful CEOs and their impact on corporate performance, Review of Financial Studies, 1443‐1432.  

Aghion, Philippe, and Jeremy Stein, 2008, Growth vs. margins: Destabilizing consequences of giving the stock market what it wants, Journal of Finance 63, 1025‐1058. 

Aharony, Joseph, and Itzhak Swary, 1980, Quarterly dividend and earnings announcements and stockholders’ returns: An empirical analysis, Journal of Finance 35, 1‐12. 

Aktas, Nihat, Eric de Bodt, Helen Bollaert, and Richard Roll, 2010, CEO narcissism and the takeover process, UCLA working paper. 

Allen, Franklin, and Gerald Faulhaber, 1989, Signaling by underpricing in the IPO market, Journal of Financial Economics 23, 303‐323. 

Alti, Aydogan, 2006, How persistent is the impact of market timing on capital structure?, Journal of Finance 61, 1681‐1710. 

Andrade, Gregor, Mark L. Mitchell, and Erik Stafford, 2001, New evidence and perspectives on mergers, Journal of Economic Perspectives 15, 103‐120. 

Ang, Andrew, Li Gu, and Yael V. Hochberg, 2007, Is IPO underperformance a peso problem?, Journal of Financial and Quantitative Analysis 42, 565‐594. 

Ang, James S., and Yingmei Cheng, 2006, Direct evidence on the market‐driven acquisition theory, Journal of Financial Research 29, 199‐216. 

Axelson Ulf, Tim Jenkinson, Per Stromberg, and Michael Weisbach, 2011, Borrow cheap, buy high? the determinants of leverage and pricing in buyouts, Working paper. 

Baker, H. Kent, and Gary E. Powell, 1993, Further evidence on managerial motives for stock splits, Quarter Journal of Business and Economics 32, 20‐31. 

Baker, Malcolm, 2009, Capital market‐driven corporate finance, Annual Review of Financial Economics 1, 181‐205. 

Baker, Malcolm, Joshua Coval, and Jeremy Stein, 2007, Corporate financing decisions when investors take the path of least resistance, Journal of Financial Economics 84, 266‐298. 

Baker, Malcolm, C. Fritz Foley, and Jeffrey Wurgler, 2009, Multinationals as arbitrageurs: The effect of stock market valuations on foreign direct investment, Review of Financial Studies 22. 337‐369. 

Page 89: Behavioral Corporate Finance: An Updated Survey · Behavioral corporate finance, and behavioral finance more broadly, received a boost from the spectacular rise and fall of Internet

  85 

Baker, Malcolm, Robin Greenwood, and Jeffrey Wurgler, 2003, The maturity of debt issues and predictable variation in bond returns, Journal of Financial Economics 70, 261‐291.  

Baker, Malcolm, Robin Greenwood, and Jeffrey Wurgler, 2009, Catering through nominal share prices, Journal of Finance 64, 2559–2590.  

Baker, Malcolm, Xin Pan, and Jeffrey Wurgler, 2011, The effect of reference point prices on mergers and acquisitions, NYU working paper. 

Baker, Malcolm, Jeremy Stein and Jeffrey Wurgler, 2003, When Does the Market Matter?, Stock Prices and the Investment of Equity‐Dependent Firms, Quarterly Journal of Economics 118, 969‐1006. 

Baker, Malcolm, and Jeremy Stein, 2004, Market liquidity as a sentiment indicator, Journal of Financial Markets 7, 271‐299. 

Baker, Malcolm, Ryan Taliaferro, and Jeffrey Wurgler, 2006, Predicting returns with managerial decision variables: Is there a small‐sample bias?, Journal of Finance 61, 1711‐1730. 

Baker, Malcolm and Jeffrey Wurgler, 2000, The equity share in new issues and aggregate stock returns, Journal of Finance 55, 2219‐2257. 

Baker, Malcolm, and Jeffrey Wurgler, 2002, Market timing and capital structure, Journal of Finance 57, 1‐32. 

Baker, Malcolm, and Jeffrey Wurgler, 2004a, A catering theory of dividends, Journal of Finance 59, 271‐288. 

Baker, Malcolm, and Jeffrey Wurgler, 2004b, Appearing and disappearing dividends: The link to catering incentives, Journal of Financial Economics 73, 271‐288. 

Baker, Malcolm, and Jeffrey Wurgler, 2011, Government bonds and the cross‐section of stock returns, NYU working paper. 

Baker, Malcolm, and Yuhai Xuan, 2011, Under new management: Equity issues and the attribution of past returns, HBS working paper. 

Barber, Brad M., and Richard Lyon, 1997, Detecting long‐run abnormal stock returns: The empirical power and specification of test statistics, Journal of Financial Economics 43, 341‐372.  

Barberis Nicholas, 2011, Psychology and the financial crisis of 2007‐2008, in: Financial Innovation and Crisis, M. Haliassos ed., forthcoming. 

Barberis, Nicholas, Ming Huang, and Richard H. Thaler, 2006, Individual preferences, monetary gambles, and stock market participation: A case for narrow framing, American Economic Review 96, 1069‐1090. 

Page 90: Behavioral Corporate Finance: An Updated Survey · Behavioral corporate finance, and behavioral finance more broadly, received a boost from the spectacular rise and fall of Internet

  86 

Barberis, Nicholas, and Andrei Shleifer, 2003, Style investing, Journal of Financial Economics 75, 283‐317. 

Barberis, Nicholas, Andrei Shleifer, and Jeffrey Wurgler, 2005, Comovement, Journal of Financial Economics 75, 283‐317. 

Barberis, Nicholas, and Richard H. Thaler, 2003, A survey of behavioral finance, in: George Constantinides, Milton Harris, Rene Stulz, eds., Handbook of the Economics of Finance, (Amsterdam: North‐Holland). 

Barberis, Nicholas, and Wei Xiong, 2009, What drives the disposition effect? An Analysis of a Long‐standing, Preference‐based Explanation, Journal of Finance 64, 751‐784. 

Barro, Robert J., 1990, The stock market and investment, Review of Financial Studies 3, 115‐132. 

Becker Bo, Zoran Ivkovic, Scott Weisbenner, 2011, Local Dividend Clienteles, Journal of Finance 66, 655‐684. 

Benartzi, Shlomo, 2001, Excessive extrapolation and the allocation of 401(k) accounts to company stock, Journal of Finance 56, 1747‐64. 

Ben‐David, Itzhak, 2010, Dividend policy decisions, in: Behavioral Finance, H. Kent Baker and John Nofsinger, eds. (Hoboken: Wiley). 

Ben‐David, Itzhak, John R. Graham, and Campbell R. Harvey, 2010, Managerial miscalibration, Duke University working paper. 

Benveniste, Lawrence, and Paul Spindt, 1989, How investment bankers determine the offer price and allocation of new issues, Journal of Financial Economics 24, 343‐362. 

Berger, Philip, and Eli Ofek, 1995, Diversification’s effect on firm value, Journal of Financial Economics 37, 39‐65. 

Bergman, Nittai K., and Dirk Jenter, 2007, Employee sentiment and stock option compensation, Journal of Financial Economics 84, 667‐712. 

Bergstresser, Daniel, Mihir Desai, and Joshua D. Rauh, 2006, Earnings manipulation, pension assumptions, and managerial investment decisions, Quarterly Journal of Economics 121, 157‐195.  

Bergstresser, Daniel, and Thomas Philippon, 2006, CEO incentives and earnings management, Journal of Financial Economics 80, 511‐529.  

Bernard, Victor, 1989, Post‐earnings announcement drift: delayed price response or risk premium? Journal of Accounting Research 27, 1‐36. 

Bertrand, Marianne, and Antoinette Schoar, 2003, Managing with style: The effect of managers on firm policies, Quarterly Journal of Economics 118, 1169‐1208. 

Page 91: Behavioral Corporate Finance: An Updated Survey · Behavioral corporate finance, and behavioral finance more broadly, received a boost from the spectacular rise and fall of Internet

  87 

Bhattacharya, Sudipto, 1979, Imperfect information, dividend policy, and the bird in the hand fallacy, Bell Journal of Economics and Management Science 10, 259‐270. 

Blanchard, Olivier, Florencio Lopez‐de‐Silanes, and Andrei Shleifer, 1994, What do firms do with cash windfalls?, Journal of Financial Economics 36, 337‐360. 

Blanchard, Olivier, Chanyong Rhee, and Lawrence Summers, 1993, The stock market, profit, and investment, Quarterly Journal of Economics 108, 115‐136. 

Bolton, Patrick, Hui Chen, and Neng Wang, 2011, Market timing, investment, and risk management, Working paper. 

Bolton, Patrick, Jose Scheinkman, and Wei Xiong, 2005, Pay for short‐term performance: Executive compensation in speculative markets, Journal of Corporation Law 30, 721‐747. 

Bouwman, Christa, Kathleen Fuller, and Amrita Nain, 2003, The performance of stock‐price driven acquisitions, University of Michigan working paper.  

Bradshaw, Mark T., Scott A. Richardson, and Richard G. Sloan, 2006, The relation between corporate financing activities, analysts’ forecasts and stock returns, Journal of Accounting and Economics 42, 53‐85.  

Brav, Alon, 2000, Inference in long‐horizon event studies: A Bayesian perspective, Journal of Finance 55, 1979‐2016. 

Brav, Alon, and Paul A. Gompers, 1997, Myth or reality?, The long‐run underperformance of initial public offerings: Evidence from venture capital and nonventure capital‐backed companies, Journal of Finance 52, 1791‐1822. 

Brav, Alon, John Graham, Campbell R. Harvey, and Roni Michaely, 2005, Payout policy in the 21st century, Journal of Financial Economics 77, 483‐528.  

Brennan, Michael J., and Thomas E. Copeland, 1988, Stock splits, stock prices, and transaction costs, Journal of Financial Economics 22, 83‐101. 

Brennan, Michael J., and Patricia J. Hughes, 1991, Stock prices and the supply of information, Journal of Finance 46, 1665‐1691. 

Brennan, Michael, and Claudia Tamarowski, 2000, Investor relations, liquidity and stock prices, Journal of Applied Corporate Finance 12, 26‐37. 

Brown, Greg, Peter Crabb, and David Haushalter, 2002, Are firms successful at selective hedging?, Journal of Business 79, 2925‐2950. 

Burch, Timothy R., William G. Christie, and Vikram Nanda, 2004, Do firms time equity offerings?, Evidence from the 1930s and 1940s, Financial Management 33, 5‐23. 

Page 92: Behavioral Corporate Finance: An Updated Survey · Behavioral corporate finance, and behavioral finance more broadly, received a boost from the spectacular rise and fall of Internet

  88 

Burgstahler, David, and Ilia D. Dichev, 1997, Earnings management to avoid earnings decreases and losses, Journal of Accounting and Economics 24, 99‐126. 

Butler, Alexander W., Gustavo Grullon, and James P. Weston, 2005, Can managers forecast aggregate market returns?, Journal of Finance 60, 963‐986.  

Camerer, Colin, and Daniel Lovallo, 1999, Overconfidence and excess entry: An experimental approach, American Economic Review 89, 306‐318. 

Camerer, Colin, and Ulrike Malmendier, 2009, Behavioral economics of organizations, in: P. Diamond and H. Vartiainen, eds., Behavioral Economics and Its Applications, (Princeton: Princeton University Press).  

Campello, Murillo, and John Graham, 2007, Do stock prices influence corporate decisions? Evidence from the technology bubble, National Bureau of Economic research working paper. 

Carslaw, C. A. P. N., 1988, Anomalies in income numbers: Evidence of goal oriented behavior, The Acconting Review 63, 321‐327. 

Chen, Long, and Xinlei Zhao, 2005, Profitability, mean reversion of leverage ratios, and capital structure choices, Washington University working paper. 

Chen, Long, and Xinlei Zhao, 2007, Mechanical mean reversion of leverage ratios, Economics Letters 95, 223‐229. 

Chernenko, Sergey, C. Fritz Foley, and Robin Greenwood, 2010, Agency costs, mispricing, and ownership structure, National Bureau of Economic Research working paper. 

Chirinko, Robert S., and Huntley Schaller, 2001, Business fixed investment and ‘bubbles’: The Japanese case, American Economic Review 91, 663‐680. 

Chirinko, Robert S., and Huntley Schaller, 2004, Glamour vs. value: The real story, Emory University working paper.  

Conlisk, John, 1996, Why bounded rationality?, Journal of Economic Literature 34, 669‐700.  

Cooper, Arnold C., Carolyn Y. Woo, and William C. Dunkelberg, 1988, Entrepreneurs’ perceived chances for success, Journal of Business Venturing 3, 97‐108. 

Cooper, Michael J., Orlin Dimitrov, and P. Raghavendra Rau, 2001, A rose.com by any other name, Journal of Finance 56, 2371‐2388. 

Cooper, Michael J., Huseyin Gulen, and P. Raghavendra Rau, 2005, Changing names with style: Mutual fund name changes and their effects on fund flows, Journal of Finance 60, 2825‐2858. 

Page 93: Behavioral Corporate Finance: An Updated Survey · Behavioral corporate finance, and behavioral finance more broadly, received a boost from the spectacular rise and fall of Internet

  89 

Cooper, Michael J., Ajay Korana, Igor Osobov, Ajay Patel, and P. Raghavendra Rau, 2005, Managerial actions in response to a market downturn: Valuation effects of name changes in the dot.com decline, Journal of Corporate Finance 11, 319‐335.  

Core, John, and Wayne Guay, 2001, Stock option plans for non‐executive employees, Journal of Financial Economics 61, 253‐287. 

Coval, Joshua D., Jakub W. Jurek, and Erik Stafford, 2009, Economic catastrophe bonds, American Economic Review 99, 628‐666. 

Cronqvist, Henrik, Anil Makhija, and Scott Yonker, 2011, Behavioral consistency in corporate finance: CEO personal and corporate leverage, Claremont McKenna College working paper. 

Dahlquist, Magnus, and Frank de Jong, 2008, Pseudo market timing: A reappraisal, Journal of Financial and Quantitative Analysis 43, 547‐580. 

Damodaran, Aswath, 2011, Applied Corporate Finance, 3rd ed. (New York: Wiley). 

De Long, J. Bradford, Andrei Shleifer, Lawrence H. Summers, and Robert Waldmann, 1989, The size and incidence of losses from noise trading, Journal of Finance 44, 681‐696. 

De Long, J. Bradford, Andrei Shleifer, Lawrence H. Summers, and Robert Waldmann, 1990, Noise trader risk in financial markets, Journal of Political Economy 98, 703‐738. 

DeAngelo , Harry, Linda DeAngelo and Douglas J. Skinner, 2004, Are dividends disappearing?, Dividend concentration and the consolidation of earnings, Journal of Financial Economics 72, 425‐456.  

DeAngelo , Harry, Linda DeAngelo and Rene Stulz, 2010, Seasoned equity offerings, market timing, and the corporate lifecycle, Journal of Financial Economics 95, 275‐295. 

Degeorge, Francois, Jayendu Patel, and Richard Zeckhauser, 1999, Earnings management to exceed thresholds, Journal of Business 72, 1‐33. 

Della Vigna, Stefano, and Ulrike Malmendier, 2004, Contract design and self‐control: Theory and evidence, Quarterly Journal of Economics 119, 353‐402. 

Della Vigna, Stefano, and Josh Pollet, 2009, Investor inattention, firm reaction, and Friday earnings announcements, Journal of Finance 64, 709‐749. 

Della Vigna, Stefano, and Josh Pollet, 2011, Capital budgeting vs. market timing: An evaluation using demographics, Journal of Finance, forthcoming. 

Derrien, Francois, 2010, Initial public offerings, in: Behavioral Finance, H. Kent Baker and John Nofsinger, eds. (Hoboken: Wiley). 

Dichev, Ilia D., 2007, What are stock investors’ actual historical returns? Evidence from dollar‐weighted returns, American Economic Review 97, 386‐402.  

Page 94: Behavioral Corporate Finance: An Updated Survey · Behavioral corporate finance, and behavioral finance more broadly, received a boost from the spectacular rise and fall of Internet

  90 

Dittman, Ingolf, and Ernst Maug, 2007, Lower salaries and no options? On the optimal structure of executive pay, Journal of Finance 62, 303‐343. 

Dittman, Ingolf, Ernst Maug, and Oliver Spalt, 2010, Sticks or carrots? Optimal CEO compensation when managers are loss averse, Journal of Finance 65, 2015‐2050. 

Donaldson, Gordon, 1961, Corporate Debt Capacity: A Study of Corporate Debt Policy and the Determination of Corporate Debt Capacity, (Boston, MA: Division of Research, Graduate School of Business Administration, Harvard University).  

Donaldson, Gordon, 1990, Voluntary restructuring: The case of General Mills, Journal of Financial Economics 27, 114‐141.  

Dong, Ming, 2010, Mergers and acquisitions, in: Behavioral Finance, H. Kent Baker and John Nofsinger, eds. (Hoboken: Wiley). 

Dong, Ming, David Hirshleifer, Scott Richardson, and Siew Hong Teoh, 2006, Does investor misvaluation drive the takeover market?, Journal of Finance 61, 725‐762.  

Dougal, Casey, Joseph Engelberg, Christopher A. Parsons, and Edward D. Van Wesep, 2011, Anchoring and the cost of capital, University of North Carolina working paper. 

Duffee, Gregory R., 1996, Idiosyncratic variation of Treasury bill yields, Journal of Finance 51, 527‐551. 

Dyl, Edward, and William B. Elliott, 2006, The share price puzzle, Journal of Business 79, 2045‐2066. 

Eckbo, B. Espen, Ronald A. Masulis, and Oyvind Norli, 2000, Seasoned public offerings: Resolution of the ‘new issues puzzle’, Journal of Financial Economics 56, 251‐292. 

Eckbo, B. Espen, and Oyvind Norli, 2004, Liquidity risk, leverage, and long‐run IPO returns, Journal of Corporate Finance 11, 1‐35.  

Edelen, Roger M., and Gregory B. Kadlec, 2005, Issuer surplus and the partial adjustment of IPO prices to public information, Journal of Financial Economics 77, 347‐373.  

Erel, Isil, Brandon Julio, Woojin Kim, and Michael S. Weisbach, 2011, Macroeconomic conditions and the structure of securities, Review of Financial Studies, forthcoming.  

Erickson, Merle, and Shiing‐wu Wang, 1999, Earnings management by acquiring firms in stock for stock mergers, Journal of Accounting and Economics 27, 149‐176. 

Fama, Eugene F., 1998, Market efficiency, long‐term returns, and behavioral finance, Journal of Financial Economics 49, 283‐306. 

Fama, Eugene F., and Kenneth R. French, 1992, The cross‐section of expected stock returns, Journal of Finance 47, 427‐465. 

Page 95: Behavioral Corporate Finance: An Updated Survey · Behavioral corporate finance, and behavioral finance more broadly, received a boost from the spectacular rise and fall of Internet

  91 

Fama, Eugene F., and Kenneth R. French, 2001, Disappearing dividends: Changing firm characteristics or lower propensity to pay?, Journal of Financial Economics 60, 3‐44. 

Fama, Eugene F., and Kenneth R. French, 2002, Testing tradeoff and pecking order predictions about dividends and debt, Review of Financial Studies 15, 1‐37.  

Fama, Eugene F., and Kenneth R. French, 2005, Financing decisions: Who issues stock?, Journal of Financial Economics 76, 549‐582. 

Fang, Lily, Victoria Ivashina, and Josh Lerner, 2010, Unstable equity? Combining banking with private equity investing, Harvard Business School working paper. 

Faulkender, Michael, 2005, Hedging or market timing?, Selecting the interest rate exposure of corporate debt, Journal of Finance 60, 931‐962.  

Ferris, Stephen P., Nilanjan Sen, and Ho Pei Yui, 2006, God save the Queen and her dividends: Corporate payouts in the UK, Journal of Business 79, 1149‐1173. 

Fischer, Stanley, and Robert C. Merton, 1984, Macroeconomics and finance: The role of the stock market, Carnegie‐Rochester Conference Series on Public Policy 21, 57‐108. 

Flannery, Mark, and Kasturi Rangan, 2006, Partial adjustment toward target capital structures, Journal of Financial Economics 79, 469‐506. 

Froot, Kenneth A., and Emil M. Dabora, 1999, How are stock prices affected by the location of trade?, Journal of Financial Economics 53, 189‐216. 

Fuller, Kathleen P., and Michael A. Goldstein, 2011, Do dividends matter more in declining markets? Journal of Corporate Finance 17, 457‐473.  

Gabaix, Xavier, 2011, A Sparsity‐Based Model of Bounded Rationality, NYU Stern School of Business working paper. 

Gao, Xiaohui, and Jay Ritter, 2010. The Marketing of Seasoned Equity Offerings. Journal of Financial Economics 97, 33‐52. 

Gaspar, José‐Miguel, Massimo Massa, and Pedro Matos, 2005, Shareholder investment horizons and the market for corporate control, Journal of Financial Economics 76, 135‐165. 

Gennaioli Nicola, Andrei Shleifer, and Robert Vishny, 2011, Neglected risks, financial innovation, and financial fragility, Journal of Financial Economics, forthcoming. 

Gervais, Simon, 2010, Capital budgeting and other investment decisions, in: Behavioral Finance, H. Kent Baker and John Nofsinger, eds. (Hoboken: Wiley). 

Gervais, Simon, J. B. Heaton, and Terrence Odean, 2010, Overconfidence, compensation contracts, and capital budgeting, Journal of Finance, forthcoming. 

Page 96: Behavioral Corporate Finance: An Updated Survey · Behavioral corporate finance, and behavioral finance more broadly, received a boost from the spectacular rise and fall of Internet

  92 

Gervais, Simon, and Terrence Odean, 2001, Learning to be overconfident, Review of Financial Studies 14, 1‐27.  

Gider, Jasmine, and Dirk Hackbarth, 2010, Financing decisions, in: Behavioral Finance, H. Kent Baker and John Nofsinger, eds. (Hoboken: Wiley). 

Gilchrist, Simon, Charles Himmelberg, and Gur Huberman, 2005, Do stock price bubbles influence corporate investment?, Journal of Monetary Economics 52, 805‐827. 

Gitman, Lawrence J., and John R. Forrester, Jr., 1977, A survey of capital budgeting techniques used by major US firms, Financial Management 6, 66‐71. 

Goel, Anand M., and Anjan V. Thakor, 2008, Overconfidence, leadership selection, and corporate governance, Journal of Finance 63, 2737‐2784.  

Golbe, Devra L., and Lawrence J. White, 1988, A time series analysis of mergers and acquisitions in the US economy, in: Alan J. Auerbach, ed., Corporate Takeovers: Causes and Consequences (NBER: University of Chicago Press). 

Gompers, Paul A., Joy Ishii, and Andrew Metrick, 2003, Corporate governance and equity prices, Quarterly Journal of Economics 118, 107‐155. 

Gompers, Paul A., and Josh Lerner, 2003, The really long‐run performance of initial public offerings: The pre‐Nasdaq evidence, Journal of Finance 58, 1355‐1392. 

Graham, John R., 1999, Duke University CFO outlook survey 1999Q2, http://www.duke.edu/~jgraham/99q2/q299ind.htm 

Graham, John R., 2000, How big are the tax benefits of debt?, Journal of Finance 55, 1901‐1941.  

Graham, John R., and Campbell R. Harvey, 2001, The theory and practice of corporate finance: Evidence from the field, Journal of Financial Economics 60, 187‐243. 

Graham, John R., Campbell R. Harvey, and Shiva Rajgopal, 2005, The economic implications of corporate financial reporting, Journal of Accounting and Economics 40, 3‐73.  

Greenwood, Robin, 2008, Excess comovement of stock returns: Evidence from cross‐sectional variation in Nikkei 225 weights, Review of Financial Studies 21, 1153‐1186. 

Greenwood, Robin, Samuel Hanson, and Jeremy C. Stein, 2010, A gap‐filling theory of corporate debt maturity choice”, Journal of Finance 65, 993‐1028. 

Greenwood, Robin, and Samuel Hanson, 2010, Issuer quality and corporate bond returns, Working paper.  

Greenwood, Robin, and Samuel Hanson, 2011, Share issuance and factor timing, Journal of Finance, forthcoming. 

Page 97: Behavioral Corporate Finance: An Updated Survey · Behavioral corporate finance, and behavioral finance more broadly, received a boost from the spectacular rise and fall of Internet

  93 

Greenwood, Robin, and Dimitri Vayanos, 2010, Price pressure in the government bond market, American Economic Review 100, 585‐590. 

Grinblatt, Mark, and Chuan Yang Hwang, 1989, Signaling and the pricing of new issues, Journal of Finance 44, 393‐420. 

Grossman, Sanford J., and Oliver D. Hart, 1988, One share‐one vote and the market for corporate control, Journal of Financial Economics 20, 175‐202. 

Guedes, Jose, and Tim Opler, 1996, The determinants of the maturity of corporate debt issues, Journal of Finance 51, 1809‐1833. 

Guedj, Ilan, and David Scharfstein, 2008, Organizational scope and investment: Evidence from the drug development strategies of biopharmaceutical firms, MIT Sloan working paper. 

Hackbarth, Dirk, 2009, Determinants of corporate borrowing: A behavioral perspective, Journal of Corporate Finance 15, 389‐411. 

Hall, Brian J., and Kevin J. Murphy, 2002, Stock options for undiversified executives, Journal of Accounting and Economics 33, 3‐42. 

Harris, Lawrence, and Eitan Gurel, 1986, Price and volume effects associated with changes in the S&P 500 list: New evidence for the existence of price pressures, Journal of Finance 41, 815‐829. 

Harris, Milton, and Artur Raviv, 1985, A sequential signaling model of convertible debt call policy, Journal of Finance 40, 1263‐1281. 

Harris, Milton, and Artur Raviv, 2008, A theory of board control and size, Review of Financial Studies 21, 1797‐1832. 

Hart, Oliver, 2008, Economica Coase lecture: Reference points and the theory of the firm, Economica 75, 404‐411. 

Hart, Oliver, and John Moore, 1990, Property rights and the nature of the firm, Journal of Political Economy 98, 1119‐1158. 

Heaton, J. B., 2002, Managerial optimism and corporate finance, Financial Management 31, 33‐45. 

Henderson, Brian J., Narasimhan Jegadeesh, and Michael S. Weisbach, 2006, World markets for raising new capital, Journal of Financial Economics 82, 63‐101. 

Hermalin, Benjamin E., and Michael S. Weisbach, 2003, Boards of directors as an endogenously determined institution: A survey of the economic literature, Economic Policy Review 9, 7‐26. 

Hietala, Pekka, Steven N. Kaplan, and David T. Robinson, 2003, What is the price of hubris? Using takeover battles to infer overpayments and synergies, Financial Management 32, 1‐32. 

Page 98: Behavioral Corporate Finance: An Updated Survey · Behavioral corporate finance, and behavioral finance more broadly, received a boost from the spectacular rise and fall of Internet

  94 

Hirshleifer, David, Angie Low, and Siew Hong Teoh, 2010, Are overconfident CEOs better innovators?, University of California at Irvine working paper. 

Hirshleifer, David and Ivo Welch, 2002, An economic approach to the psychology of change: Amnesia, inertia, and impulsiveness, Journal of Economics and Management Strategy 11, 379‐421.  

Holmstrom, Bengt, 1982, Managerial incentive problems – a dynamic perspective, republished in 1999, Review of Economic Studies 66, 169‐82. 

Hong, Harrison, Jiang Wang, and Jailin Yu, 2008, Firms as buyers of last resort, Journal of Financial Economics 88, 119‐145. 

Hovakimian, Armen, 2006, Are observed capital structures determined by equity market timing?, Journal of Financial and Quantitative Analysis 41, 221‐243.  

Hovakimian, Armen, Tim Opler, and Sheridan Titman, 2001, The debt‐equity choice, Journal of Financial and Quantitative Analysis 36, 1‐24. 

Hu, Xing, Jun Pan, and Jiang Wang, 2011, Noise as Information for Illiquidity, Princeton University working paper. 

Huang, Rongbing, and Jay Ritter, 2009, Testing theories of capital structure and estimating the speed of adjustment, Journal of Financial and Quantitative Analysis 44, 237‐271. 

Huddart, Steven, Mark Lang, and Michelle H. Yetman, 2009, “Volume and price patterns around a stock’s 52‐week highs and lows: Theory and evidence,” Management Science 55, 16‐31. 

Ikenberry, David, Josef Lakonishok, and Theo Vermaelen, 1995, Market underreaction to open market share repurchases, Journal of Financial Economics 39, 181‐208. 

Ikenberry, David, Josef Lakonishok, and Theo Vermaelen, 2000, Stock repurchases in Canada: Performance and strategic trading, Journal of Finance 55, 2373‐2398. 

Jain, Bharat A., and Omesh Kini, 1994, The post‐issue operating performance of IPO firms, Journal of Finance 49, 1699‐1726. 

Jensen, Michael C., 1986, Agency costs of free‐cash‐flow, corporate finance, and takeovers, American Economic Review 76, 323‐329. 

Jensen, Michael C., 2005, Agency costs of overvalued equity, Financial Management 34, 5‐19.  

Jensen, Michael C., and William Meckling, 1976, Theory of the firm: Managerial behavior, agency costs and ownership structure, Journal of Financial Economics 3, 305‐360. 

Jensen, Michael C., and Richard S. Ruback, 1983, The market for corporate control: The scientific evidence, Journal of Financial Economics 11, 5‐50. 

Page 99: Behavioral Corporate Finance: An Updated Survey · Behavioral corporate finance, and behavioral finance more broadly, received a boost from the spectacular rise and fall of Internet

  95 

Jenter, Dirk, 2005, Market timing and managerial portfolio decisions, Journal of Finance 60, 1903‐1949. 

John, Kose, and Joseph Williams, 1985, Dividends, dilution, and taxes: A signalling equilibrium, Journal of Finance 40, 1053‐1070. 

Jung, Kooyul, Yong Cheol Kim, and Rene M. Stulz, 1996, Timing, investment opportunities, managerial discretion, and the security issue decision, Journal of Financial Economics 42, 159‐185. 

Kahneman, Daniel, 2003, Maps of bounded rationality: Psychology for behavioral economics, American Economic Review 93, 1449‐1475. 

Kahneman, Daniel, and Amos Tversky, 1979, Prospect theory: An analysis of decision under risk, Econometrica 47, 263‐291. 

Kaplan, Steven N., Mark Klebanov, and Morten Sorenson, 2011, Which CEO characteristics and abilities matter?, Journal of Finance, forthcoming. 

Kaplan, Steven N., and Per Stromberg, 2003, Financial contracting theory meets the real world: An empirical analysis of venture capital contracts, Review of Economic Studies 70, 281‐315. 

Kaplan, Steven N., and Michael Weisbach, 1992, The success of acquisitions: Evidence from divestitures, Journal of Finance 47, 107‐138. 

Kaustia, Markku, 2004, Market‐wide impact of the disposition effect: Evidence from IPO trading volume, Journal of Financial Markets 7, 207‐235. 

Kaustia, Markku, 2010a, Disposition effect, in: Behavioral Finance: Investors, Corporations, and Markets, H. Kent Baker and John R. Nofsinger, eds., (Hoboken, NJ: Wiley) 

Kaustia, Markku, 2010b, Prospect theory and the disposition effect, Journal of Financial and Quantitative Analysis 45, 791‐812. 

Kayhan, Ayla, and Sheridan Titman, 2004, Firms’ histories and their capital structures, NBER working paper no. 10526. 

Keynes, John Maynard, 1936, The General Theory of Employment, Interest, and Money, (London: Macmillan). 

 Kindleberger, Charles, 2000, Manias, Panics, and Crashes, (New York: Wiley). 

Klein, Peter G., 2001, Were the acquisitive conglomerates inefficient?, RAND Journal of Economics 32, 745‐761. 

Kolasinski, Adam, and Xu Li, 2010, Do independent directors restrain overconfidence CEOs?, University of Washington working paper. 

 

Page 100: Behavioral Corporate Finance: An Updated Survey · Behavioral corporate finance, and behavioral finance more broadly, received a boost from the spectacular rise and fall of Internet

  96 

Kothari, S. P., and Jay Shanken, 1997, Book‐to‐market, dividend yield, and expected market returns: A time‐series analysis, Journal of Financial Economics 44, 169‐203. 

 Krishnamurphy, Arvind, 2002, The bond/old bond spread, Journal of Financial Economics 66, 463‐

506.  Kruger, Philipp, Augustin Landier, and David Thesmar, 2011, The WACC fallacy: The real effects of 

using a unique discount rate, Toulouse School of Economics working paper. 

La Porta, Rafael, 1996, Expectations and the cross‐section of stock returns, Journal of Finance 51, 1715‐1742. 

La Porta, Rafael, Josef Lakonishok, Andrei Shleifer, and Robert Vishny, 1997, Good news for value stocks: Further evidence on market efficiency, Journal of Finance 52, 859‐874. 

Lakonishok, Josef, Andrei Shleifer, and Robert W. Vishny, 1994, Contrarian investment, extrapolation, and risk, Journal of Finance 49, 1541‐1578.  

Lamont, Owen, 2000, Investment plans and stock returns, Journal of Finance 55, 2719‐2745. 

Lamont, Owen, and Jeremy Stein, 2006. Investor sentiment and corporate finance: Micro and macro, American Economic Review 96(2), 147‐151 

Landier, Augustin, and David Thesmar, 2009, Financial contracting with optimistic entrepreneurs, Review of Financial Studies 22, 117‐150.  

Lang, Larry, and Rene Stulz, 1994, Tobin’s q, corporate diversification and firm performance, Journal of Political Economy 102, 1248‐1280. 

Langer, Ellen J., and Jane Roth, 1975, Heads I win tails it’s chance: The illusion of control as a function of the sequence of outcomes in a purely chance task, Journal of Personality and Social Psychology 32, 951‐955. 

Larwood, Laurie, and William Whittaker, 1977, Managerial myopia: Self‐serving biases in organizational planning, Journal of Applied Psychology 62, 194‐198. 

Leary, Mark, and Michael R. Roberts, 2005, Do firms rebalance their capital structures?, Journal of Finance 60, 2575‐2619. 

Leland, Hayne, and David Pyle, 1977, Informational asymmetries, financial structure, and financial intermediation, Journal of Finance 32, 371‐387. 

Lerner, Joshua, 1994, Venture capitalists and the decision to go public, Journal of Financial Economics 35, 293‐316. 

Li, Wei, and Erik Lie, 2006, Dividend changes and catering incentives, Journal of Financial Economics 80, 293‐308. 

Page 101: Behavioral Corporate Finance: An Updated Survey · Behavioral corporate finance, and behavioral finance more broadly, received a boost from the spectacular rise and fall of Internet

  97 

Lintner, John, 1956, Distribution of incomes of corporations among dividends, retained earnings and taxes, American Economic Review 46, 97‐113. 

Ljungqvist, Alexander, Vikram Nanda, and Rajdeep Singh, 2006, Hot markets, investor sentiment, and IPO pricing, Journal of Business 79, 1667‐1702. 

Ljungqvist, Alexander and William Wilhelm, 2005, Does prospect theory explain IPO market behavior?, Journal of Finance 60, 1759‐1790.  

Long, John, 1978, The market valuation of cash dividends: A case to consider, Journal of Financial Economics 6, 235‐264. 

Loughran, Tim, Jay Ritter, and Kristian Rydqvist, 1994, Initial public offerings: International insights, Pacific‐Basin Finance Journal 2, 165‐199. 

Loughran, Tim, and Jay Ritter, 1995, The new issues puzzle, Journal of Finance 50, 23‐51. 

Loughran, Tim, and Jay Ritter, 1997, The operating performance of firms conducting seasoned equity offerings, Journal of Finance 52, 1823‐1850. 

Loughran, Tim, and Jay Ritter, 2000, Uniformly least powerful tests of market efficiency, Journal of Financial Economics 55, 361‐389.  

Loughran, Tim, and Jay Ritter, 2002, Why don’t issuers get upset about leaving money on the table in IPOs?, Review of Financial Studies 15, 413‐443.  

Loughran, Tim, and Anand M. Vijh, 1997, Do long‐term shareholders benefit from corporate acquisitions?, Journal of Finance 52, 1765‐90. 

Lyon, Richard, Brad M. Barber, and Chih‐Ling Tsai, 1999, Improved methods for tests of long‐run       abnormal stock returns, Journal of Finance 54, 165‐201.  

Malkiel, Burton, 1990, A Random Walk Down Wall Street, (New York: W.W. Norton). 

Malmendier, Ulrike, and Geoffrey Tate, 2005, CEO overconfidence and corporate investment, Journal of Finance 60, 2661‐2700.  

Malmendier, Ulrike, and Geoffrey Tate, 2008, Who makes acquisitions? CEO overconfidence and the market’s reaction, Journal of Financial Economics 89, 20‐43.  

Manove, Michael, and A. Jorge Padilla, 1999, Banking (conservatively) with optimists, RAND Journal of Economics 30, 324‐350. 

March, James G., and Zur Shapira, 1987, Managerial perspectives on risk and risk taking, Management Science 33, 1404‐1418. 

Marsh, Paul, 1982, The choice between equity and debt: An empirical study, Journal of Finance 37, 121‐44. 

Page 102: Behavioral Corporate Finance: An Updated Survey · Behavioral corporate finance, and behavioral finance more broadly, received a boost from the spectacular rise and fall of Internet

  98 

Massa, Massimo, Urs Peyer, and Zhenxu Tong, 2005, Limits of arbitrage and corporate financial policy, INSEAD working paper. 

Matsusaka, John G., 1993, Takeover motives during the conglomerate merger wave, RAND Journal of Economics 24, 357‐379. 

McNichols, Maureen F., and Stephen R. Stubben, 2008, Does earnings management affect firms’ investment decisions? The Accounting Review 83, 1571‐1603. 

Merrow, Edward W., Kenneth E. Phillips, and Christopher W. Myers, 1981, Understanding cost growth and performance shortfalls in pioneer process plants, (Santa Monica, CA: Rand). 

Mikkelson, Wayne H., Megan Partch, and Ken Shah, 1997, Ownership and operating performance of companies that go public, Journal of Financial Economics 44, 281‐307. 

Miller, Edward M., 1977, Risk, uncertainty, and divergence of opinion, Journal of Finance 32, 1151‐1168. 

Miller, Merton H., and Kevin Rock, 1985, Dividend policy under asymmetric information, Journal of Finance 40, 1031‐1051. 

Mitchell, Mark L. and Erik Stafford, 2000, Managerial decisions and long‐term stock price performance, Journal of Business 73, 287‐330. 

Modigliani, Franco, and Merton H. Miller, 1958, The cost of capital, corporation finance, and the theory of investment, American Economic Review 48, 655‐669. 

Moeller, Sara, Frederik Schlingemann, and René Stulz, 2005, Wealth destruction on a massive scale?, A study of acquiring‐firm returns in the recent merger wave, Journal of Finance 60, 757‐782. 

Morck, Randall, 2010, Loyalty, agency conflicts, and corporate governance, in: Behavioral Finance, H. Kent Baker and John Nofsinger, eds. (Hoboken: Wiley). 

Morck, Randall, Andrei Shleifer, and Robert W. Vishny, 1990a, Do managerial objectives drive bad acquisitions?, Journal of Finance 45, 31‐48. 

Morck, Randall, Andrei Shleifer, and Robert W. Vishny, 1990b, The stock market and investment: Is the market a sideshow?, Brookings Papers on Economic Activity 2:1990, 157‐215. 

Morck, Randall, and Fan Yang, 2002, The mysterious growing value of S&P 500 membership, University of Alberta working paper. 

Moskowitz, Tobias J., and Annette Vissing‐Jorgensen, 2002, The returns to entrepreneurial investment: A private equity premium puzzle?, American Economic Review 92, 745‐778.  

Muelbroek, Lisa, 1992, An empirical analysis of illegal insider trading, Journal of Finance 47, 1661‐1699. 

Page 103: Behavioral Corporate Finance: An Updated Survey · Behavioral corporate finance, and behavioral finance more broadly, received a boost from the spectacular rise and fall of Internet

  99 

Myers, Stewart C., 1977, Determinants of corporate borrowing, Journal of Financial Economics 5, 147‐175. 

Myers, Stewart C., and Nicholas S. Majluf, 1984, Corporate financing and investment decisions when firms have information that investors do not have, Journal of Financial Economics 13, 187‐221. 

Ofek, Eli and Matthew Richardson, 2003, DotCom Mania: The rise and fall of Internet stocks, Journal of Finance 58, 1113‐1138. 

Oyer, Paul, and Scott Schaeffer, 2005, Why do some firms give stock to all employees?, An empirical examination of alternative theories, Journal of Financial Economics 76, 99‐133. 

Pagano, Marco, Fabio Panetta, and Luigi Zingales, 1998, Why do companies go public?, An empirical analysis, Journal of Finance 53, 27‐64. 

Panageas, Stavros, 2003, Speculation, overpricing, and investment: Theory and empirical evidence, MIT working paper.  

Polk, Christopher, and Paola Sapienza, 2009, The stock market and corporate investment: A test of catering theory, Review of Financial Studies 2009, 187‐217.  

Pontiff, Jeffrey, 1996, Costly arbitrage: Evidence from closed‐end funds, Quarterly Journal of Economics 111, 1135‐1151. 

 Pontiff, Jeffrey, and Lawrence D. Schall, 1998, Book‐to‐market ratios as predictors of market 

returns, Journal of Financial Economics 49, 141‐160.  Porter, Michael E., 1987, From competitive advantage to corporate strategy, Harvard Business 

Review 65, 43‐59.  Rajan, Raghuram, 1992, Insiders and outsiders: The choice between informed and arm’s‐length 

debt, Journal of Finance 47, 1367‐1400.  

Rau, P. Raghavendra, and Theo Vermaelen, 1998, Glamour, value and the post‐acquisition performance of acquiring firms, Journal of Financial Economics 49, 223‐253. 

Ravenscraft, David J., and Frederic M. Scherer, 1987, Mergers, Sell‐Offs, and Economic Efficiency, (Washington, DC: Brookings Institution).  

Rhodes‐Kropf, Matthew, David Robinson, and S. Viswanathan, 2005, Valuation waves and merger activity: The empirical evidence, Journal of Financial Economics 77, 561‐604.  

Rhodes‐Kropf, Matthew, and S. Viswanathan, 2004, Market valuation and merger waves, Journal of Finance 59, 2685‐2718.  

Page 104: Behavioral Corporate Finance: An Updated Survey · Behavioral corporate finance, and behavioral finance more broadly, received a boost from the spectacular rise and fall of Internet

  100 

Richardson, Scott A., and Richard G. Sloan, 2003, External financing and future stock returns, University of Pennsylvania working paper. 

Ritter, Jay, 1991, The long‐run performance of initial public offerings, Journal of Finance 42, 365‐394. 

Ritter, Jay, 2003, Investment banking and securities issuance, in: George Constantinides, Milton Harris, Rene Stulz, eds., Handbook of the Economics of Finance, (Amsterdam: North‐Holland). 

Rosch, E. H., 1973, Natural categories, Cognitive Psychology 4, 328‐350. 

Roll, Richard, 1986, The hubris hypothesis of corporate takeovers, Journal of Business 59, 197‐216. 

Ross, Stephen, 1977, The determination of financial structure, Bell Journal of Economics 8, 23‐40. 

Savor, Pavel, and Qi Lu, 2009, Do stock mergers create value for acquirers?, Journal of Finance 64, 1059‐1095. 

Scarpetta, Stefano, Philip Hemmings, Thierry Tressel, and Jaejoon Woo, 2002, The role of policy and institutions for productivity and firm dynamics: Evidence from micro and industry data, OECD working paper. 

Schaller, Huntley, 1993, Asymmetric information, liquidity constraints and Canadian investment, Canadian Journal of Economics 26, 552‐574. 

Schaller, Huntley, 2011, The economic effect of sentiment, Carleton University working paper.  

Scharfstein, David, and Jeremy Stein, 1990, Herd behavior and investment, American Economic Review 80, 465‐479. 

Scheinkman, Jose, and Wei Xiong, 2003, Overconfidence and speculative bubbles, Journal of Political Economy 111, 1183‐1219. 

Schneider, Christoph, and Oliver Spalt, 2010, Acquisitions as lotteries: Do managerial gambling attitudes influence takeover decisions? University of Mannheim working paper.  

Schultz, Paul, 2003, Pseudo market timing and the long‐run underperformance of IPOs, Journal of Finance 58, 483‐517. 

Schultz, Paul, 2004, Pseudo market timing and the stationarity of the event‐generating process, University of Notre Dame working paper. 

Sen, Rik, and Robert Tumarkin, 2010, Stocking up: Executive optimism and share retention, Hong Kong University of Science and Technology working paper. 

Seyhun, H. Nejat, 1992, Why does aggregate insider trading predict future stock returns?, Quarterly Journal of Economics 107, 1303‐31. 

Page 105: Behavioral Corporate Finance: An Updated Survey · Behavioral corporate finance, and behavioral finance more broadly, received a boost from the spectacular rise and fall of Internet

  101 

Shefrin, Hersh, 2000, Beyond Greed and Fear, (Boston, MA: Harvard Business School Press). 

Shefrin, Hersh, 2001, Behavioral Corporate Finance, Journal of Applied Corporate Finance 14, 113‐124. 

Shefrin, Hersh, 2006, Behavioral Corporate Finance, (New York: McGraw‐Hill). 

Shefrin, Hersh, 2008, Ending the Management Illusion, (New York: McGraw‐Hill). 

Shefrin, Hersh and Meir Statman, 1984, Explaining investor preference for cash dividends, Journal of Financial Economics 13, 253‐282. 

Shefrin, Hersh, and Meir Statman, 1985, The disposition to sell winners too early and ride losers too long: Theory and evidence, Journal of Finance 40, 770‐790. 

Shleifer, Andrei, 1986, Do demand curves for stocks slope down?, Journal of Finance 41, 579‐590. 

Shleifer, Andrei, 2000, Inefficient markets: An introduction to behavioral finance, (Oxford, UK: Oxford University Press). 

Shleifer, Andrei, and Robert Vishny, 1997, The limits of arbitrage, Journal of Finance 52, 35‐56. 

Shleifer, Andrei, and Robert Vishny, 2003, Stock market driven acquisitions, Journal of Financial Economics 70, 295‐312. 

Shleifer, Andrei, and Robert Vishny, 2010, Unstable banking, Journal of Financial Economics 97, 306‐318. 

Simon, Herbert A., 1955, A behavioral model of rational choice, Quarterly Journal of Economics 69, 99‐118. 

Skinner, Douglas J., and Richard G. Sloan, 2002, Earnings surprises, growth expectations, and stock returns or don’t let an earnings torpedo sink your portfolio, Review of Accounting Studies 7, 289‐312. 

Sloan, Richard, 1996, Do stock prices fully reflect information in accruals and cash flows about future earnings?, The Accounting Review 71, 289‐316. 

Spence, A. Michael, 1973, Job market signaling, Quarterly Journal of Economics 87, 355‐379. 

Spiess, Katherine, and John Affleck‐Graves, 1995, Underperformance in long‐run stock returns following seasoned equity offerings, Journal of Financial Economics 38, 243‐267. 

Spiess, Katherine, and John Affleck‐Graves, 1999, The long‐run performance of stock returns following debt offerings, Journal of Financial Economics 54, 45‐73. 

Stambaugh, Robert F., 1999, Predictive regressions, Journal of Financial Economics 54, 375‐421. 

Page 106: Behavioral Corporate Finance: An Updated Survey · Behavioral corporate finance, and behavioral finance more broadly, received a boost from the spectacular rise and fall of Internet

  102 

Statman, Meir, and James F. Sepe, 1989, Project termination announcements and the market value of the firm, Financial Management 18, 74‐81. 

Statman, Meir, and Tyzoon T. Tyebjee, 1985, Optimistic capital budgeting forecasts: An experiment, Financial Management 14, 27‐33. 

Stein, Jeremy C., 1989, Efficient capital markets, inefficient firms: a model of myopic corporate behavior, Quarterly Journal of Economics 104, 655‐669. 

Stein, Jeremy C., 1996, Rational capital budgeting in an irrational world, Journal of Business 69, 429‐455. 

Stein, Jeremy C., 2003, Agency, information and corporate investment, in: George Constantinides, Milton Harris, Rene Stulz, eds., Handbook of the Economics of Finance, (Amsterdam: North‐Holland).  

Stigler, George J., 1964, Public regulation of the securities markets, Journal of Business 37, 117‐142. 

Strack, Fritz, and Thomas Mussweiler, 1997, Explaining the enigmatic anchoring effect: Mechanisms of selective accessibility, Journal of Personality and Social Psychology 73, 437‐446. 

Svenson, Ola, 1981, Are we all less risky and more skillful than our fellow drivers?, Acta Psychologica 47, 143‐148.  

Teoh, Siew Hong, Ivo Welch, and T. J. Wong, 1998a, Earnings management and the long‐run market performance of initial public offerings, Journal of Finance 53, 1935‐74. 

Teoh, Siew Hong, Ivo Welch, and T. J. Wong, 1998b, Earnings management and the underperformance of seasoned equity offerings, Journal of Financial Economics 50, 63‐99. 

Thaler, Richard H., 1980, Towards a positive theory of consumer choice, Journal of Economic Behavior and Organization, 1, 39‐60. 

Thaler, Richard H., 1985, Mental accounting and consumer choice, Marketing Science 4, 199‐214. 

Thomas, Jacob, 1989, Unusual patterns in reported earnings, The Accounting Review 64, 773‐787. 

Titman, Sheridan, K.C. John Wei, and Feixue Xie, 2004, Capital investments and stock returns, Journal of Financial and Quantitative Analysis 39, 677‐700. 

Townsend, Richard R.,2011, Propagation of financial shocks: The case of venture capital, Dartmouth Tuck School of Business working paper.  

Tversky, Amos, and Daniel Kahneman, 1974, Judgment under uncertainty: Heuristics and biases, Science 185, 1124‐1130. 

Tversky, Amos, and Daniel Kahneman, 1991, Loss aversion in riskless choice: A reference‐dependent model, Quarterly Journal of Economics 106, 1039‐1061. 

Page 107: Behavioral Corporate Finance: An Updated Survey · Behavioral corporate finance, and behavioral finance more broadly, received a boost from the spectacular rise and fall of Internet

  103 

Viswanathan, S., and Bin Wei, 2008, Endogenous events and long run returns, Review of Financial Studies 21, 855‐888. 

Weinstein, Neil D., 1980, Unrealistic optimism about future life events, Journal of Personality and Social Psychology 39, 806‐820.  

Welch, Ivo, 1989, Seasoned offerings, imitation costs, and the underpricing of initial public offerings, Journal of Finance 44, 421‐449. 

Welch, Ivo, 2004, Stock returns and capital structure, Journal of Political Economy 112, 106‐131. 

Welch, Ivo, 2009, Corporate Finance: An Introduction, (New Jersey: Prentice). 

Weld, William C., Roni Michaely, Richard H. Thaler and Shlomo Benartzi, 2009, The nominal price puzzle, Journal of Economic Perspectives 23(2), 121‐42. 

Wurgler, Jeffrey, 2011, On the economic consequences of index‐linked investing, in: W. T. Allen, R. Khurana, J. Lorsch, and G. Rosenfeld, eds., Challenges to Business in the Twenty‐First Century: The Way Forward (Cambridge: American Academy of Arts and Sciences). 

Wurgler, Jeffrey, and Ekaterina Zuravskaya, 2002, Does arbitrage flatten demand curves for stocks?, Journal of Business 75, 583‐608. 

Yermack, David, 1996, Higher market valuation of companies with a small board of directors, Journal of Financial Economics 40, 185‐211. 

Zhang, Donghang, 2004, Why do IPO underwriters all allocate extra shares when they expect to buy them back? Journal of Financial and Quantitative Analysis 39, 571‐594.